SCCM Resource Library-Death and Dying in the ICU: Bringing Lessons From ICU Liberation to End-of-Life ICU Care

Video Transcription

Hello, my name is Randall Curtis. I'm Director of the Cambia Palliative Care Center of Excellence  at the University of Washington, and I'm going to talk today about the question, what is high-quality end-of-life care in the ICU, to kick off this session. I have no financial  conflicts of interest to disclose. I do like to acknowledge our funding sources, including our healthcare system, the Cambia Health Foundation, the NIH, and PCORI. So what I  thought I would do is answer these two questions. First, what is high-quality end-of-life care  in the ICU? And second, can interventions increase this quality? In terms of what high-quality  is, I think that we can see some observational data to suggest that we have room for improvement. These are data from Quill and colleagues, published in 2014, looking at variability  in withholding and withdrawing life support in the U.S. across different ICUs. Each dot here represents a different ICU, and this looks at the proportion of deaths that are  preceded by a decision to withhold or withdraw life-sustaining treatments. And what we see is tremendous variability from ICU to ICU, even after adjusting for severity of illness,  and ICU characteristics. This degree of variability suggests that there is important opportunity to reach consensus in the way we approach these issues and improve care.  We conducted a systematic review looking at variability across the world in studies that have been published, mostly from Europe and North America, but other regions as well.  And specifically, we divided these studies looking at variability in withholding and withdrawing life support into three groups. Studies that examine variability between countries,  studies that examine variability within ICUs in the same region or same country, and then even one study looking at variability between physicians in a single ICU. And I show these  data to make the point that although there's important variability between countries and physicians, there's also variability within countries and even within a single ICU. Again,  suggesting opportunities for improvement. This is an important paper published in JAMA looking at the perception of appropriateness of care amongst ICU nurses and physicians.  This cross-sectional study across 82 ICUs surveying over 1,600 clinicians in Europe and Israel found that clinicians reported delivering inappropriate care for at least  one patient 27% of the time. And if you look at the reasons behind inappropriate care,  you see a number of reasons that raise concern about the quality of care. A lack of participation  in decision-making by patients and families, inaccurate information to patients and families, not good quality care, and disregarding patients' wishes.  If we think of the effect that this poor quality care could have on family members, I think  it's important to highlight that we know that family caregivers of critically ill patients have important psychological sequelae. This is a recent systematic review of 40 studies  showing the prevalence of psychological symptoms very widely, but much higher than in the general  public. Depression from 20 to 94%, anxiety from 20 to 80%, and PTSD from 15 to 62% in most of the studies. And in fact, the longer the patient was in the ICU was associated  with increased risk of PTSD, anxiety, and depression in some of these studies. And a patient dying in the ICU, not surprisingly, has been associated with increased symptoms  of PTSD and depression. These studies suggest that one important sequelae of poor quality care in the ICU may be increased distress amongst family caregivers.  Another way to think about quality is to look at families' ratings of quality. And this is an important and interesting study that was published recently looking at family surveys  from over 28,000 patients in 102 VA hospitals in the US. And what this study looked at was family ratings of quality depending on how much time the patient spent in the ICU during  their terminal hospitalization. And the interesting finding here is that the more, the higher the proportion of time the patient spent in the ICU, the higher the family's ratings of  quality of end-of-life care. And so I think this is important to understand as we think about quality of care in the ICU where we have more nurses and perhaps more attention  to family, we actually see higher family ratings of quality. So I'd argue that high quality end-of-life care in the ICU is multifaceted. As we think  about improving quality, we need to focus on improving patient and family-centered outcomes, including psychological symptoms, quality of communication, and quality of the care  experience. We also need to strive to provide goal-concordant care, that is, match the intensity  of care to informed and supported patient goals. This is a key outcome, but one that's  very difficult to measure. We need to be good stewards of ICU resources and provide appropriate care. And then importantly, variability in care should be based on patient and family  values, not on physician values. And in fact, I think one of the major advances in thinking  about improving palliative end-of-life care in the ICU is to think broader than just end-of-life  care. If we define palliative care as care focused on improving communication about goals of care, maximizing comfort and quality of life, end-of-life care, care for patients  who are actively dying is an important part of that, but it's only a part. And I think as long as we're focusing on end-of-life care, we miss important opportunities to improve  palliative care for patients who are dying, but we don't recognize them as dying, or patients who have unmet palliative care needs, but are not actively dying.  So can we improve quality? I'm going to show several studies that I think offer proof of concept that focusing on the way we provide care in the ICU can in fact improve quality.  And we can think about that in terms of different models for integrating palliative and critical  care. One is to increase palliative care skills of ICU clinicians, what's often called primary  palliative care. A second is to incorporate palliative care specialists or other specialists into the ICU, especially palliative care. And then what I would argue is probably the  model that makes the most sense is to incorporate both, training in palliative care for ICU clinicians and palliative care consultation. I think this is an important study talking  about proof of concept, published in New England Journal in 2007 from France. This study randomized  126 patients that the attending physician thought would die within a few days. The intervention  was to have a proactive family conference on the day of randomization or the following day using the value strategy, and I'll show that, as well as giving a bereavement pamphlet  to the family. Here's a value strategy, five steps to explicitly acknowledge family statements,  acknowledge family emotions, listen to the family, understand the patient as a person, and elicit family questions. And this study looked at family member outcomes in terms  of symptoms of anxiety, depression, and PTSD at three months. Two important findings, one very high level of symptoms in the control group, and second, this relatively simple  intervention focusing on improving communication in the ICU with family members had dramatic improvements in family member symptoms of anxiety, depression, and PTSD.  A second trial I'll point to was published in American Journal of Respiratory and Critical Care Medicine in 2016. This was our trial of a communication facilitator. We identified  critically ill patients with respiratory failure who were randomized to intervention or usual care.  The intervention was this communication facilitator, a nurse or a social worker, who was trained to facilitate communication between the ICU team and the family to address  individual communication needs for family members and to identify, address, and mediate conflict.  We enrolled 268 family members of 168 patients, and what we found was a trend towards reduced depression at three months, significant reduction in depression at six months,  no change in anxiety, and a trend towards reduced PTSD at six months.  We also found reduced ICU length of stay amongst patients that died, reduced costs, particularly amongst patients that died, and then reduced average daily costs, suggesting  more targeted use of ICU resources, making decisions earlier when patients ultimately are not able to survive the ICU stay. Another study that I think demonstrates proof of concept  is this study by Doug White and colleagues, a randomized trial of family support intervention.  This was a stepped wedge cluster randomized trial with a multi-component family support  intervention that was delivered by the ICU team and led by trained ICU nurses or partner nurses conducted in five ICUs with over 1,400 patients.  The outcomes they looked at included anxiety, depression, and PTSD. They did not see a significant improvement there. They did see a significant improvement in families' ratings  of the quality of communication as well as family ratings of the patient-centeredness of care. They also importantly saw reduced healthcare utilization with decreased length  of ICU stay and decreased length of hospital stay with the intervention.  I think there are a number of lessons that we've learned about what high-quality care is, end-of-life  care is in the ICU. Importantly, I believe it is multifaceted. I think it's very important that  we focus on palliative and not end-of-life care so we don't miss opportunities to improve care  for patients and their families. Primary palliative care interventions can work, and I've showed you  some of those, but it's important that we understand and adapt to the local culture, including the ICU culture. Especially palliative care interventions can also work  but must be integrated with the ICU team. Ultimately, I believe in clinical practice,  the best model is the mixed model, including both primary and specialty palliative care interventions.  I think that is an important way forward is to figure out how best to accomplish those goals.  With that, I'll stop, and thank you very much for your attention.  Thank you so much to the conference organizers for this opportunity to speak with you today.  I have no relevant financial disclosures, but I would like to start by acknowledging all of the  individuals who directly and indirectly participated in the work that I'll be presenting today, specifically my co-investigators, who are all named here, as well as the patients,  the clinicians who participated in the ICU Liberation Collaborative. I think this slide  will be a surprise to no one, which is to say that ICU admission is common at the end-of-life, particularly in the United States, where approximately one in five adults who die in  the U.S. spend time in an ICU at or near the end-of-life. This translates to over 500,000 individuals who receive end-of-life care in an ICU each year in the United States.  This recognition that providing end-of-life care is an integral part of our profession in critical care medicine is what motivated this study that I'm going to be presenting,  which sought to describe both the quality and the variability of end-of-life care that is  currently delivered within ICUs in the United States. Today, with the time I have, I'm going to be focusing on unit-level variation in end-of-life care delivery. Our work has been  recently published in JAMA Network Open. To accomplish our study objectives, we conducted  an observational cohort study within the framework of a large nationwide collaborative sponsored by  the Society of Critical Care Medicine and the Gordon and Betty Moore Foundation. As many of  you know, the ICU Liberation Learning Collaborative sought to disseminate the A-through-F care bundle,  which is listed here, which primarily focuses on the management of pain, agitation, and delirium in the ICU. The collaborative organizers were generous enough to allow us to design  within the infrastructure of the collaborative a prospective cohort study looking at end-of-life  care delivery. The ICU Liberation was comprised of 68 intensive care units, as you can see here,  distributed across the United States and one in Puerto Rico that represented medical, surgical,  mixed specialty units and were from a range of urban, suburban, and rural environments.  Our study focused on the over 1,500 patients who died within the ICU during the learning  collaborative or ICU decedents. As you can see, 42 percent of these decedents were over the age of  80, and the race distribution mirrors that of the U.S. Census. What I'd like to present to you today are findings on the unit-level variation in performance in five key end-of-life  events. We chose to measure these five specific events because prior observational work has found  that these five events have all been associated with higher quality ratings of death and dying  by family members and ICU clinicians. In other words, these five events are meaningful and,  in general, at the population level, represent favorable end-of-life events. I'm going to show  you now a series of graphs or results that all look fairly similar. Each dot on this plot  represents one of the 68 ICUs in the collaborative. The size of the dot represents the relative number  of decedents from that unit. On the y-axis is the proportion of patients within that ICU who experienced one of the five end-of-life events. The dotted line is the median event rate  across the entire collaborative. Starting, what you can see here is comparing between ICUs.  There's a wide variation in the number, the proportion of patients who were extubated prior  to death among those that were mechanically ventilated. Similarly, we saw notable differences  between units in the number, the proportion of patients who had absence of CPR in the last  hour of life. When we go on to look at symptom control, we again saw notable differences between  ICUs. First, looking at the proportion of patients who were delirium-free in the last 24 hours of  life and who were pain-free again in the last 24 hours of life. Finally, we also found important  differences between units in the proportion of patients who had family or significant other  persons present at the time of death. When we put all of these plots together and take a step  back, what you find is a picture of wide unit-level variability in end-of-life care delivery  among USICUs who participated in the ICU Liberation Collaborative. For our research team, the next natural question that these data generated was whether it was the same ICUs  at the higher performing end for each end-of-life event or were there other patterns of care delivery  and trade-offs within end-of-life care delivery. To answer this question, we pursued an analysis known as k-means clustering. This is an unsupervised machine learning technique  that seeks to identify groups of ICUs, or it clusters the ICUs together based on similar patterns of end-of-life care delivery. One of the benefits about this type of clustering is  that we did not apply a priori hypotheses about what we thought would potentially be a pattern  of care. Instead, the patterns emerged from the data themselves. What you're seeing here on the screen are cluster plots where we tested different numbers of clusters of ICUs. What  you can see here is that we were able to group the ICUs from the ICU Liberation Collaborative into  three mutually exclusive and similar clusters of end-of-life care delivery. Next, what I'll show  you are the patterns of care represented by each cluster. What you can see here represented  along each of these five end-of-life events is the first cluster. This was 24 of the 68 units. On the y-axis, again, is the proportion of patients from that cluster who experienced  the favorable end-of-life event. What you can see by this cluster is that, in general, this cluster of units performed highly in each of the five different end-of-life events  that we measured. The second cluster had similar performance in four of the domains, but you can see had the lowest performance rate in delirium-free performance. Then, finally, the  third cluster had the lowest number of patients who were extubated prior to death and the lowest  number of patients who avoided CPR. In other words, when we step back, what this characterizes  is three different patterns of care delivery with one cluster achieving very high performance  across all five domains. Our study has some limitations, notably that it was housed within  an existing ICU Liberation Collaborative, which is both a strength, but also it was not designed  to represent or to study end-of-life care delivery and is not a random sample of ICUs.  Second, the high-quality end-of-life care and the most important end-of-life events lack  clear definition. We chose our study measures and events based on expert opinion, ethical principles,  and limited observational data. In summary, we identified wide variation in end-of-life care delivery between US ICUs. We found three mutually exclusive unit-level patterns of care  delivery that included high-performing units. We believe that future research should target this unit-level variation and further study and then disseminate the successes of these  high-performing ICUs. Thank you very much, and I look forward to questions and further discussion.  I'd like to thank the organizing committee for inviting me to speak today, and today I'll be  talking about the Ethicus 2 study, Trends in End-of-Life Care Over Time, a European Perspective,  which was published last year in JAMA. I have no conflicts of interest. If we look at the Ethicus  1 study, which was published in JAMA in 2003, we can see differences between northern, central,  and southern Europe in terms of the end-of-life categories. We can see a lot more withdrawing than in the north than in the south, and a lot more CPR in the south than in the north.  Now, since that study, there have been new laws, recommendations, and guidelines for end-of-life  practices, greater awareness of end-of-life issues, palliative care has improved in Europe and worldwide, and more communication tools for end-of-life decision-making,  greater physician knowledge about outcomes, and more consensus for end-of-life practices.  The hypothesis of the Ethicus 2 study was that there's been a change in end-of-life practices over the last 16 years from the time of the Ethicus 1 study. We repeated the  prospective observational study to evaluate changes. The study was from September 2015  to September 2016, and patients were enrolled for a self-selected, continuous six-month period.  Consecutively admitted ICU patients who either died or who had life-saving therapy limitations  were prospectively studied. They were followed up until ICU discharge, death for two months  from the first decision to limit life-prolonging therapies, and we used the same definitions and  methodologies as we used in our first Ethicus study. The senior intensivists completed the study data form, and 22 of the original 37 Ethicus ICUs participated in this study.  We used the same end-of-life category definitions as we used before, withholding treatment,  not to start or increase a life-sustaining intervention, withdrawing treatment to actively  stop a life-sustaining intervention presently being given, active shortening of the dying process,  and act with the specific intent of shortening the dying process, CPR, death despite ventilation  and cardiac massage, and brain death documented cessation of cerebral function and meeting criteria for brain death. Now if we look at the results of the study,  here we have Ethicus I, the first study, in blue, and the Ethicus II study in red, we can see that over time we have much less CPR, moving from 22 percent to 6.2 percent,  and much more withholding, 41 percent to 50 percent, and more withdrawing, 25 percent to 39 percent, and these differences were statistically significant. Now if we look at the same  end-of-life categories by region, the north, central, and southern regions, we can see a  decrease in CPR in all three of the regions as seen here. We can see an increase of withholding  and withdrawing in the north, and the same increase in withholding and withdrawing central,  and also in the south, an increase in withholding and withdrawing. Despite these changes, we still  see differences between the south and mostly the north. You can see that we have more withdrawing  in the south now, but when you compare the withdrawing in the south compared to the central  or to the northern areas, it's still much less withdrawing in the south. And if we look at all the limitations, withholding, withdrawing, and active shortening of the dying process,  we can see that over time we've had an increase in the number of limitations. When we look at the outcomes based on these limitations, we can see that the overall mortality in here,  I guess one in orange, efficacy two in blue, we see a decrease in the mortality.  The survival after withholding therapy was statistically significant and a much higher survival, 5% after withholding in efficacy two, but in withdrawing we have a significant  increase in survival. And this was not just withholding CPR, it's occurred after withholding vasopressors and mechanical ventilation and also withdrawing of them. And survival after  all the limitations was significantly higher in efficacy two, approximately one-fifth of the  patients. When we look at the various timeframes, we can see that the ICU length of stay were  shorter in efficacy two. The time from ICU admission to the first limitation was statistically  much less, almost half the time in efficacy two. And the time from the first treatment limitation until death was longer in efficacy two, but not statistically significant,  where it was statistically significant for a longer time from withholding treatment until death and a statistically significant shorter time from withdrawing treatment until death.  Now, to try to understand these reasons for limitation, we performed a logistic regression  analysis, controlling for potential confounders that you can see here, see why we were limiting  versus not limiting treatments. Interesting in this study, we performed an ethical practice score,  which was a retrospective evaluation of 12 variables, routine family meetings, daily  deliberation, appropriate levels of care, end-of-life discussion during meetings, written triggers for limitation, written end-of-life guidelines and protocols,  palliative care or ethics consultations, staff taking communication or bioethics courses, and country end-of-life guidelines or legislation. And when we look at the results of this  multivariable analysis, we can see that the statistical significance of the variables were found for time, we had more limitations as we saw in efficacy two, for age, for physician  religion, chronic diseases, but the greatest odds ratio, interestingly enough, was for the  ethical practice score. So in conclusion, end-of-life practice have changed considerably  in 16 years. Patients with treatment limitations or dying in 2013 or 16 had significantly more frequent limitations in life prolonging therapies and significantly less frequent  death without limitation, CPR or brain death, compared to the same ICUs in 1999-2000.  Less CPR was provided before death, more frequent and earlier treatment limitations, and active shortening of the dying process still remains very uncommon  in the ICU. More patients with life-prolonging therapy limitations survived.  More regional differences persisted despite the various changes that we saw. So despite life-sustaining treatments being limited more frequently and earlier, more patients  survived the hospital discharge, suggesting that limiting life-sustaining treatments should not  automatically be considered life-ending events. I'd especially like to thank the investigators  from the 22 ICUs who were very helpful in making this study possible, and I thank you very much for your attention. Bye.  Welcome, everybody, and thank you so much for joining us in this session, Death and Dying in the ICU, bringing lessons from the ICU Liberation Collaborative to end-of-life  care. My name is Jackie Kruiser, and I'm an assistant professor of medicine in the Division  of Allergy, Pulmonary, and Critical Care at the University of Wisconsin, and I have the pleasure of moderating  today's session with Dr. Wes Ely, who is a professor of medicine in the Grant W. Little Chair in Medicine at Vanderbilt University Medical Center in Nashville, where he's also  director of the Critical Illness, Brain Dysfunction, and Survivorship Center. We also have a distinguished  panel of speakers and discussants here today. So you heard first earlier from Dr. Randy Curtis  from the University of Washington, where he's a professor of medicine and director of the  Cambia Palliative Care Center of Excellence. And also you have heard from Dr. Charlie Sprump, who is director emeritus in the Department of Anesthesia and Critical Care Medicine  at Hadassah Hebrew University Medical Center in Jerusalem. And next you will actually have  the pleasure of hearing a few words live from Dr. Judy Nelson, who's joining us from Memorial Sloan-Kettering, where she is chief of the Supportive Care Service. And Judy's going to  talk to us a little bit about the future of end-of-life care delivery in the ICU.  Thank you so much, Jackie. And I hope you'll bear with me all without slides that were not able to  be uploaded. But I do just want to then make a short discussion of a few key points. And I think,  you know, envisioning can be defined in two ways. One is what we expect or believe is going to  happen. And the other is, you know, what we imagine as future possibilities. And I think  it's more liberating to let ourselves imagine what it might be like. And it frees us to really  appreciate even radical changes that we would have to make to do things better. I think we can  look into the future also with the lens of COVID. Here we are in New York City, closely approaching  a year from the beginning of this awful pandemic, which has brought challenges for end-of-life care  in the ICU, really unlike any that I think we anticipated. And they're still going on, and they're quite extreme. And we can learn from them. The first point I want to make is that  it has been our tradition to want to know what is going to happen. We have many very  sophisticated tools to predict those things. And yet, we really haven't gotten to where I suppose  we hoped we would be, where we could know who is going to die in the ICU and who won't, so that we could somehow or other focus our end-of-life care on the ones who had the worst  odds and focus our intensive care on the ones who we thought would survive. And it turns out  that we're still not very good at that. And I think we can say for the future that we will  continue to have uncertainty. And what does that mean for us? Well, we certainly learned this  during COVID, right, that we had a disease that we had never seen before. We didn't know what was  going to happen. And we had to face it on those terms. And what it means most to me is that the  emotional and spiritual work that we have to do at the end of life with our patients, their families,  and for ourselves, for that matter, is going to continue to be unavoidable. And it's something  that we can improve if we focus on it in those terms. We can't just know. We have to deal with  the uncertainty. And we have to wonder about who our patients are and what psychosocial and economic  context they're coming from so that we can provide them the best end-of-life care. The second point I  would make is that we have focused a lot on how we tell patients and families in the ICU information  that they need to make what Randy called informed and supported decisions, which I think is a  beautiful way of expressing goal-concordant care. I think we haven't yet spent enough time on how  much we listen ourselves. We've worked a lot on addressing emotions in patients and families so  that they can hear what we say to them. But I don't think we've yet optimized how well we can hear  what they have to tell us, both as listeners in a conversation, but also as receiving information  through other means and leveraging our information technology to learn more about our patients as  people. And in a better future for end-of-life care in the ICU, we will do more of that.  Visiting restrictions are not new in COVID. This is how things were in the early days of intensive  care units. And in many places, visitor restrictions continued right up until the time of COVID,  although they became more draconian. And I can accept the fact that when we didn't have enough  PPE and things were very chaotic at the beginning of COVID, that perhaps this made sense. I'm curious  though about how long we have continued these severe restrictions, even after the PPE is more  plentiful and even after we have learned in the hardest way how traumatizing it is for everyone to exclude the family visitors and other loved ones from the intensive care unit.  It has forced our groups to become family members, which is a job that is very hard for us to take on.  And we have not begun to see the trauma of what this means for our patients' families and others.  A fourth point I'd like to make out of five is that we still have a very doctor-centric hierarchical  model in our intensive care units. And yet what we've seen in COVID, again, and in states now,  is that it is the nurses, for the most part, who are there with our patients. And throughout a lot  of COVID, they've been almost the only people in the rooms with those patients, and sometimes taking  their orders from doctors who were not even trained in critical care medicine, because we didn't have  enough, right? The shortage was worse for intensivists, much worse than it was for ventilators.  And yet these nurses, even as recently as December when I made rounds in an ICU, these nurses were outside the circle of trainees who stood around me, and those nurses were looking  at the backs of the doctors. And so we have to do a better job of disassembling this hierarchy  and really creating a team, a team that includes nurses and a team that includes chaplains, before  they are called to provide mass rights. These are people who are very important in the emotional and  spiritual work. And finally, we have been very focused on mortality at specific time points,  and I recently saw data sets from a study that had various mortality endpoints, not just in the ICU,  but in the hospital and at several post-ICU standpoint time points up to six months. And for  this intervention, mortality was higher at every single time point up to six months. This was a not true study that Randy showed of Doug White, where the six-month mortality finally  evened out, but the other post-discharge mortalities were higher in that study too. And it was said to me about the six-month difference in mortality, well,  your work shows how difficult it is to be on prolonged mechanical ventilation after the ICU,  so can we look at this outcome of higher mortality at six months and say, we did them a favor,  we spared them the post-ICU months of that kind of suffering. That is a conclusion that I'm not  prepared to accept right now. I was very disturbed by it, particularly since it refers to my own work,  and I think we're going to need to get a much deeper understanding of what the experience  looks like after the ICU, particularly when we do optimize the care in those months for patients,  and then we can evaluate what that experience is like for them. So those are the main points that  I want to make, and we can talk in the discussion about how we might try to get toward a future  vision that we share through changes in our priorities, our tools, our training, our mentorship,  and so forth. So thanks for letting me introduce the discussion in this way.  Thank you so much, Judy. That was really a great way of wrapping up the lessons that we already  learned from Jackie and Charlie and Randy, and I am going to start by telling you a quick story  of two patients, two families. The Stevens were a husband and wife who got COVID, both of them,  and were admitted into my COVID unit, and they were put in separate rooms. They've been married  for 64 years or something like that, and I went up to Mr. Stevens the next morning, and he was  looking all around for his wife. He was trying to get out of bed. He was a fall risk,  and I said, what are you doing? What are you doing? T was his name, Tom, and he said, and I have  permission to use their names, by the way. This is not a HIPAA violation. I've got signed, written  permission to use these people's names, and their pictures have been published already in Lancet  last week, Lancet Respiratory Medicine, that Lancet had a COVID frontline essay contest,  and I entered these stories in there, and it was published. So no HIPAA violation. Anyway, he said,  where's my wife? I can't find her. Anyway, long story short, we got them in the same room  next to one another. His delirium resolved, and they spent the rest of the time in the hospital  convalescing, holding hands, mattress to mattress. It was beautiful. It reminds me of what Randy said  earlier about the importance of the difference and overlap between end-of-life care and palliative  care. Maybe he can comment a little bit more about that in a few minutes, but before they were  together, to use the words that you were using, Judy, traumatizing and draconian. We had an in-hospital visitation problem for two people who had been married for 65 years.  Contrast that with this next couple, both of the Hills, both got COVID. He was a heart transplant  patient, so he got transferred to Vanderbilt, while his wife gets admitted to a rural hospital  two hours away. The daughter says to the doctor, this is crazy, they have to be together, and the  doctor says back this, well, it's not going to make a difference to their medical care, so what's the  point? She goes, he's going to die anyway, probably, and so is she, and she said, if they're going to  die anyway, then that's the only point is to put them together. They've been married for over 60  years, and so through this laborious process, we saved a room next to him, she gets transferred, and  they finally got to put together in the same room, mattress to mattress, died holding hands in the  ICU of COVID, both of them, with the family around them in PPE, because as we said earlier, PPE works.  Once you gown up, glove up, mask up, and put a shield on, that's how we're staying safe.  So why are we still using these draconian family visitation laws and rules? Because they were needed  in April, they aren't necessarily needed now, and I invoke that we should question this. I'd love to  know what other panelists say, and not every ICU has the PPE, but for those who do, if we can keep  ourselves safe, we can teach the family members how to stay safe too, and I'll close with this  comment. Pull up our COVID-D paper. The COVID-D paper was published in Lancet Respiratory last  two weeks ago. 2,000 patients, first author Brenda Pun, 2,000 COVID-ICU patients. Guess what the rate  of F was in the ABCDF bundle, family visitation? 17 percent, and that includes virtual visitation.  In-person or virtual, 17 percent. I think that is absurd. I think it's cruel. Families aren't a luxury. They are part of the treatment plan. That's the point. That's the point.  Without the family, they can't get treated right, and healing won't occur. Thoughts, comments from the panel?  Randy, Judy, Charlie, anybody about that? Should family be treated as a luxury, or are they part of the ICU treatment plan?  Well, you're preaching to the choir, Wes, but I certainly would agree with you that they are part  of the ICU treatment plan, that we really can't provide good quality care, patient-centered care,  without being inclusive of family. I think that, you know, in terms of envisioning the future, I think  we all share this view here, but not everybody shares it, right? And that was really the point  that I was trying to make, is that I think this speaks to how uncomfortable some people still are  with having visiting, open visiting the ICU, but I think for the future, what we need to think about  is not just opening the door, which should be basic, but what do we actually do with the families when they're there,  and how do we integrate them in ways that make the care really better? How do we conduct rounds with  them, for example? We, you know, some people thought about that, but we haven't really studied that on  a massive scale, and it's really quite challenging and interesting. What kind of care processes can  they participate in? So, I think, you know, with an interprofessional group that includes nursing and others,  those are the ways that we could really make the future better. I think opening a door is, you know,  almost a no-brainer at this point, but what we do is something we could challenge ourselves with.  Jackie, can you build on what Judy just said, and tell us what, in the ICU liberation program,  that the SCCM sponsored, we studied, you know, 16,000 people and published it, and the F is part of the A2F bundle.  So, what did we do in the A2F bundle in the ICU liberation program with the F?  Yeah, well, and I guess what I, one thing that I'd like to say about that, and sort of build on,  and maybe come back and ask Charlie a question here at the end, is that, you know, in our study,  where we look just at the decedents, so those patients who died in the ICU, actually, the end-of-life care event  that was most frequently achieved was having family present at the time of death. Now, certainly,  that isn't exactly what we're talking about in terms of family presence in the ICU, but nevertheless,  it's probably sort of an extreme marker of whether or not families sort of are available and integrated  in care. But, and, you know, and I was sort of, you know, interested in, perhaps, Charlie's perspective on this,  because I think, you know, I was recently talking to someone, one of our colleagues in Germany,  who was sort of shocked at the level of family access that we had in our ICUs in the United States pre-COVID,  and actually, we were reviewing the results of this paper, and she was pretty impressed by the  high rate of family presence, particularly at the end-of-life, that she felt like that was a unique feature of the United States,  or perhaps not represented worldwide. So I think, you know, I'd be curious to hear from Charlie about your perspective  in the variation at this, like, this opening to families, the sort of, the desire to have that, and sort of what the practices were pre-COVID in other places.  So I would say that pre-COVID, certainly, the family was an integral part of the team taking care of the patient.  For the patients who were conscious, but even those who were not, it was certainly very, very helpful.  I think we also have to be cognizant of the different cultures, and in Israel, particularly,  I think we also have to be cognizant of the different cultures, and in Israel, particularly,  there were families who just don't listen to the rules, and we could never really have an open unit anytime.  We occasionally would have family members come on rounds, but it culturally was impossible, because I myself, being the director of the unit, would tell families,  we have to make rounds and take care of your loved ones, and they sort of looked at me and said, who are you, anyhow? I mean, we'll stay as long as we want.  So, I mean, after you go through that a number of times, I mean, you know, it's difficult sometimes to enforce the rules  when you like to be more flexible, but when it really hurts patient care, then you can't do it.  So, I think there is a lot to do with the cultures, and I assume most places in the United States, the families listen to the nurses and doctors when they're asked to leave,  as opposed to other places around the world where that not may actually happen. But, Charlie, I started to beat on this right away, but does it really hurt care?  I mean, in our ICU, we have the families there 24-7 open visitation on rounds.  The way that our A2F bundle works is that the family is in these circles on rounds, up to two people on rounds,  and we give them a lay summary at the end of our presentation so they can hear what we said, and then we give them 30 seconds to a minute to ask a question or two,  and if they need more than that, we will go to the family conference. But that has become our usual care pre-COVID, and now we've reopened the non-COVID ICU to have that.  So, our ICU does that family visitation again. In the COVID unit, they only come between 1 and 3 in the afternoon. We need to expand that. Well, here's the thing, right?  I mean, I think there are certainly challenges involved, but your experience and others, you know, we haven't shared this enough, and that could be really helpful, right,  to collaborate around this and figure out, you know, if you have a strategy for making it work, can it work in other places?  This is not about, you know, bilting or blaming people for what has been. This is a long tradition in ICU. I just think COVID gives us an opportunity now to re-look at it, you know,  and what's been going on in COVID. In fairness, most units or a lot of units will allow a family member in if somebody is imminently dying, right?  So, you end up arguing that somebody, you know, either is or isn't dying enough to allow a family member in there.  So, you know, we're not there, but I think we can help each other find ways to make something work in any culture.  I also think a really important point that Charlie's comment brings up is that open visitation is not visitation without rules. Correct.  And, Wes, you even said yourself that, you know, open visitation 24-7,  two family members present during rounds. They can ask one or two questions during rounds.  And we have to help the nurses, the junior doctors, others enforce those rules in a way that's consistent and sensitive and not apply differently to different families.  And I think those aspects of it are really important as well. Okay. I'm going to have to move on. There's 15 questions in the chat.  I don't see them, but let's start giving the questions. We've got 10 minutes left to do the 15 questions. Yeah.  So, I'm going to, yeah, bring up, and it looks like, you know, some of the questions that have been voted to the top are focused on the,  some of the system level issues around palliative care. And so, I think, you know, any of our panelists here would be, we would love to hear your insights on this.  So, the first one from Heidi Engel is, what is your opinion of having a system of automatic palliative care consults for patients in the ICU  who are there for more than one or two weeks? Do you have examples of institutions that have instituted this?  And if the family doesn't want palliative care in this system, would palliative care providers still check in as part of the protocol?  So, maybe we'll have Randy start with that since he sort of introduced this concept of palliative care, and then certainly everybody else can join the conversation. Sure.  So, I think that it can be very helpful to have some routine triggers to consider palliative care consult.  But what I will say about that is that in a well-worked, in an ICU with high efficiency primary palliative care and access to specially palliative care for difficult cases,  I personally don't think very many automatic triggers are helpful. A lot of these automatic triggers are based on high mortality, high risk of death.  They're not necessarily based on high palliative care needs. And you can have some patients at very high risk of death with where the patient and family actually have very low,  relatively low palliative care needs that can be met by a primary palliative care team. So, I feel like where primary palliative care is working well and efficient,  these kinds of triggers tend to be less helpful. Where primary palliative care is more rudimentary,  that's where personally I've seen these triggers be particularly helpful to get palliative care involved in helping patients, families, and the ICU clinicians in these situations.  In our ICU, we only have one automatic trigger for palliative care involvement, and that's putting a patient on ECMO. And we have continued that.  Although even there, palliative care comes, checks in. They may sign off. They may follow at a distance depending on the palliative care needs.  Wonderful. And I think, you know, there is sort of a couple more questions along the same line.  So, I'll maybe summarize them here together, and we can continue this idea about palliative care.  You know, one question asking, is having palliative care in the emergency department too soon? Or perhaps, you know, sort of getting at this idea of when exactly to start.  And, you know, a second question asking about, hold on, I have a question. Okay, here we go.  So, how do you train future ICU attendings to listen and have these conversations? In your own practice, how soon do you start these conversations?  And what do you do when all the services in the ICU are not on the same page? For example, a neurologist saying there's a 1% chance, and the family latch onto that,  as opposed to listening to the ICU team. How do you approach? So, maybe the summary question I would pose to the group, and maybe we could have Judy answer this one,  is kind of when is the best time to have palliative care involved? When is too early? And how do we manage conflict in the big team?  I think Randy made the key point, which is palliative care is not based on prognosis. Right? Hospice benefit is unfortunately based on prognosis, which tends to restrict it.  But palliative care is based on need.  And unfortunately, most of the triggers that have been developed are based on prognosis. And I want to point out, this has not had a good impact on the field of palliative care,  because it tends to entrench the idea that palliative care is for dying patients. Right? So, I think if people really want triggers, they ought to focus on what they view as needs.  I think also the idea that palliative care, you know, should be mandatory, should prompt the discussion, what is the problem that you're trying to address here?  Is there a lot of distress around the care of very sick patients in the ICU being experienced by the caregivers? And if there is, how about surfacing that discussion, as opposed to,  or a concern that perhaps the ICU is not educated or trained well enough to take care of the issues. And if that's the problem, then it could be addressed on those terms,  as opposed to mandating a palliative care consultation, which almost never works well when it's forced. Right?  Either with the primary clinicians involved or with the patients and families. So, I would urge people to, you know, think about need, which also cuts across all stages of disease.  I work in a cancer center, and all the cancer centers have tortured themselves with this idea that, you know, how do we define advanced cancer so that, you know,  that's the time that you should have a conversation about health care or whatever. It makes no sense, right? I mean, there's all kinds of transitions in cancer.  A diagnosis of cancer is terrifying for a woman who has early stage breast cancer that will almost certainly be cured. So, again, we can't divide up, right, based on prognosis.  We are going to go by need, and that can be evaluated at any stage. Jackie, just a quick anecdote. I work at a hospital.  One of my hospitals I work at had really, really bad use of palliative care in the ICU.  And so, for a transient period of time, we made a mandatory palliative care consult for anyone on a ventilator, and we only needed it for about six months.  The culture shifted greatly, and now we no longer have a mandatory palliative care consult. And yet, I think we do a great job with palliative care involvement,  but the culture was so antiquated in that particular hospital that we needed to do something like that to kind of break the glass ceiling, and then it worked.  So there may be audience members who are in a situation like that where it could work for you to switch things through such a mandatory role. I think that's a fair point, Wes.  I really do. And, in fact, these treatments were developed back in the early 2000s when nobody knew what palliative care teams were like,  and they were designed for that very purpose, to show people what a palliative care consult could do.  I think there are other ways to get palliative care teams in there, like inviting them on rounds with you a couple times a week  and letting them speak to individual cases without mandating their consultation. So there are different ways to approach introducing them, but if that's the issue, you're right.  It's important for people to know what value a palliative care team can bring. I think I do want to interrupt and make a clarifying point.  Someone had asked a question, and I'm going to just sort of say the answer here.  Someone had asked, what's the difference between primary palliative care and specialty palliative care?  So to clarify for the audience, so what that refers to is primary palliative care are intensivists,  or someone in their sort of native environment, providing palliative care to patients,  or using the principles of palliative care to provide palliative care to patients and meet their needs in that way.  And specialty palliative care refers to consultation from a specialty palliative care service of specialized trained.  So normally people commonly would refer to that as a palliative care consult versus providing palliative care within your practice,  the same as you're caring for the critically ill patient. So just to clarify there.  And then I do want to, one question that's become very popular, sort of switching gears a little bit in the last couple of minutes,  is actually asking for, in most of the A-F bundle, and there's sort of this tension that pediatrics is learning from adults for the A-F bundle,  but when it comes to F, pediatrics may have a lot to offer as it's a foundation of the pediatric practice.  So, and then another question, have adult intensivists considering shadowing pediatric intensivists to see how family-centered rounds can work?  And so just kind of interested to hear everybody's perspective on how we can learn as adult intensivists. I think all of us here are from our pediatric group.  Charlie, you have a comment? Yeah, I think an important one is palliative care is extremely well developed in North America.  There are many countries that do not have any palliative care at all, certainly not teams. There are some that are just beginning to develop.  And many intensivists, even though not trained officially in palliative care, can provide palliative care.  And in many countries, some of the palliative care people just can't get to all the needs.  And as much as we'd like to move palliative care, as Judy mentioned, to any patient that needs it,  no matter what stage of their disease, they're just seeing it for end-of-life care. And there's not enough of them to go around just for that. Well, we have 45 seconds left.  I would like Jackie to close us up here. But I do want to thank the pediatric intensivists on here for reminding us how much you have to offer us in this area. I love that idea.  And I think I'm going to go on rounds in the pediatric ICU soon and learn from you. Wonderful time together, all of us.  And Judy and Charlie and Randy, thanks for sharing with us so much of your wisdom over the years. Jackie, bring it home. Yeah.  And just in addition to thanking the panelists, thanks to all of you who are here listening to us today and for engaging so nicely and asking a lot of great questions.  You know, I think this just is a testament to how, you know, our field has really come to see end-of-life  and palliative care as two distinct things but both key parts of who we are as critical care medicine clinicians. So thanks for contributing to that, everybody.  And I hope you all take care. Bye, everybody. Thank you. Bye-bye. Have a good day.

Video Summary

The panel discussed various aspects of high-quality end-of-life care in the ICU. They reviewed observational data that suggested room for improvement in end-of-life care practices, such as variability in withholding and withdrawing life support across different ICUs. They also highlighted the perception of appropriateness of care among clinicians, which revealed instances of delivering inappropriate care. The panel discussed the impact of poor quality care on family caregivers, which can lead to psychological symptoms such as depression, anxiety, and PTSD. They also emphasized the importance of family ratings of care quality and the need for patient and family-centered outcomes. The panel explored different models for integrating palliative care into the ICU, including primary palliative care, incorporation of palliative care specialists, and a mixed model with training in palliative care for ICU clinicians and palliative care consultation. They presented several studies that demonstrated proof of concept that these models can improve quality of care. The panel stressed the importance of focusing on improving communication, providing goal-concordant care, being good stewards of ICU resources, and basing variability in care on patient and family values. They concluded that high-quality end-of-life care in the ICU is multifaceted and requires collaboration among healthcare professionals to ensure patient and family-centered care.

Asset Subtitle

Ethics End of Life, Patient and Family Support, 2021

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2021 Critical Care Congress held virtually from January 31-February 12, 2021.

Learning Objectives

Describe the key domains that define high-quality end-of-life ICU care
Apply existing end-of-life ICU care quality measures to practice and compare to national and international trends
Identify at least two key strategies that will help improve end-of-life care for patients in the ICU

What Is High-Quality End-of-Life Care?
J. Randall Curtis

Variability of End-of-Life Care in U.S. ICUs: Insights From the ICU Liberation Collaborative
Jacqueline Kruser

Trends in End-of-Life Care Over Time: A European Perspective
Charles L. Sprung

Meta Tag

Content Type Presentation

Knowledge Area Ethics End of Life

Knowledge Area Patient and Family Support

Knowledge Level Foundational

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Palliative Care

Tag Brain Death

Tag Communication

Year 2021

Keywords

high-quality end-of-life care

ICU

observational data

withholding and withdrawing life support

appropriateness of care

family caregivers

psychological symptoms

palliative care

communication

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.