Hello, everyone. I am thrilled to welcome you to the ACCM Town Hall, Critical Care in Times of Crisis, Adaptation and Innovation. I'm Laurie Schutter from University of Pittsburgh. I'm the Vice Chair of Education here, and I am also on the Board of Regents and helped coordinate this session and serving as the moderator. I have the honor to introduce our speakers. Douglas White will be speaking on Resource Allocation and Guiding Difficult Decisions. Dr. White is Vice Chair and Professor of Critical Care Medicine at the University of Pittsburgh School of Medicine. He is an expert on end-of-life decision-making for patients with acute life-threatening illnesses, and he has been actively involved in developing the guidelines for resource allocation during our pandemic crisis. We will hear from Brenda Punn, who is from Vanderbilt University Medical Center. She's an advanced practice nurse with special interest in critical care. She serves as the Director of Data Quality at Vanderbilt. She's involved in a variety of research projects that focus on improving the care and outcomes of critically ill patients and their families. She will be speaking to us on patient and family engagement support during absence. So with that, I'm going to turn it over to our speakers, and thank you for joining us. Again, we will have plenty of time for questions at the end. Hello, my name is Doug White. I'm Vice Chair and Professor in the Department of Critical Care Medicine at the University of Pittsburgh School of Medicine. First, thanks so much to the organizers for inviting me to speak today. So today I'll be speaking on the topic of mitigating inequities and saving lives with ICU triage during the COVID-19 pandemic. And by way of disclosures, I receive grant funding from the NIH and also receive personal honoraria from UpToDate as a contributor, as well as from the American Journal of Respiratory and Critical Care Medicine as an associate editor. So I'll cover three main things today. First, I'll argue against the use of two proposed approaches to ICU triage that I'll call the save the most lives triage approach and equal chances triage. Second, I'll make the case that ICU triage policies should be designed to simultaneously promote population health outcomes and mitigate health inequities. And third, I'll present strategies to promote health equity during ICU triage. A couple topics I won't be addressing are the issue of re-triage and also operational aspects of training and deploying triage teams. So for many intensivists in the world, really the first signal of how bad things were going to get during the pandemic came from what played out in Lombardy, Italy. And you'll remember Lombardy is a wealthy and well-resourced area of Italy. Hospitals there were nonetheless overwhelmed. And in a JAMA podcast or webcast early in the pandemic, Maurizio Ciccone, who's the current president of ESICM described already full ICUs being confronted with 20 or 30 new patients with acute respiratory failure requiring mechanical ventilation every day. So take a moment and put yourself in the situation of the doctors in those hospitals and think about how should you decide among those patients who should be offered mechanical ventilation when not all can. And when you consider that question, if you're in the United States or another country in which there are significant health disparities, it's also critical to consider this contextual reality that persons of color are dying at two to three times the rate of white patients. And importantly, there's really no evidence that these differences arise from, for example, different genetic susceptibilities. And instead they're arising from what we call the social determinants of health. For example, persons of color are much more likely to become infected with COVID and much more likely to die from it if they contract it because of things like they contract it more because of higher housing density, a reliance on public transportation, more likely to work in essential worker positions, and also higher risk for severe infection because of the comorbidities that disproportionately affect this population. And this is not simply, these disparities don't simply arise solely due to race-based differences. It's also about socioeconomic disadvantage. And these are data from New York City showing that individuals from high poverty neighborhoods are dying at more than twice the rate of individuals from wealthy neighborhoods, again, due to what we think of as the social determinants of health. So how should we organize ICU triage when there are simply not enough beds or ventilators for all patients in need? Let's think about two options. Option one is what's been called pure utilitarian triage in that the goal here is really to allocate scarce ICU beds and ventilators to do the greatest good for the greatest number. And there are really two triage criteria often embedded in this, saving the most lives, which would manifest as using something like a SOFA score and giving priority to patients with the greatest chance of surviving the hospital discharge. And then another criterion might be saving the most lives. And here you might prioritize individuals who have the greatest life expectancy if they survive their COVID illness. These are two ways of contextualizing or conceptualizing doing the greatest good for the greatest number. And one policy that really embodies this ethic is guidance from the Canadian Medical Association, which states that in the context of a pandemic, the value of maximizing benefits is most important. Priority for limited resources should aim at both saving the most lives and maximizing improvements in individuals post-treatment length of life. All right. The problem is that the save the most lives triage strategy would almost certainly deprioritize persons of color and the poor and therefore raises really serious equity concerns. So let's look at each criterion individually. The problem with a save life years approach to triage is that individuals who are poor and persons of color have shorter life expectancies. And I should say individuals with certain disabilities do as well. And these differences really arise from social determinants of health, like poverty, poor healthcare access, food insecurity, all of which lead to higher rates of life shortening comorbid conditions like end stage renal disease, hypertension, diabetes, and heart disease. Now, what's the problem with using something like the SOFA score for chances of survival to hospital discharge? Well, mounting evidence suggests that there are higher SOFA scores on presentation among black patients compared to white patients. And there are a variety of possible causes of this, none of which, again, are due to underlying genetic susceptibility to sepsis, et cetera. But instead from things like these individuals carry a higher burden of chronic organ dysfunction, they may delay presentation to care either because of distrust of the healthcare system or poor transportation and access. And also there's increasing evidence that the kidney score of the SOFA, for example, is biased due to how it treats creatinine. So focusing solely on saving the most lives or life years with triage will likely compound the disparities in COVID-19 outcomes among persons of color, as well as economically disadvantaged groups, because these issues arise from entrenched structural racism and other forms of structural inequities. All right, so what about option two here, which is what I call equal chances for all. And what this option would entail is essentially for all eligible patients using a coin flip or other random selection process in the face of scarcity to determine who receives mechanical ventilation and who receives alternatively comfort focused care. And again, note here that prognosis would be simply irrelevant as would all other factors. It would just be random allocation of those, for example, in need of a ventilator. And this is an approach that's not theoretical. Diego Silva recently argued in favor of this approach in an article published in Chest. He wrote that all else being equal, we should maximize the number of lives saved. However, there's the rub. If the safety pre COVID-19 was not equal, nor is it during this pandemic, because doctors can't rectify existing injustices, but can likely exacerbate them through maximizing the number of lives saved approach, then randomness is preferable to efficiency in allocating ventilators. But there are at least two major concerns with this approach that make it a very poor choice, in my opinion. In short, equality in chances will lead to both inefficiency first, it will result in many preventable deaths, including among the disadvantaged groups that we're trying to help. Because patients with a poor prognosis, for example, if someone, you know, extremely likely to die, despite intensive care gets an equal chance for treatment as patients with a very good prognosis if they receive intensive care. And second, it's important to note this approach won't overcome health disparities. The disparities that we're seeing in COVID-19 arise upstream from the ICU due to social determinants of health. So random allocation will only bake in the pre existing disparities. This highlights that there's an important distinction between giving everyone equal chances and showing equal respect. And so from a justice and fairness perspective, a better way to manifest equal respect for all members of society is to take steps to mitigate the circumstances that cause already disadvantaged groups to bear the greatest burden of the pandemic. All right. So it in the last portion of this talk, I'm going to talk, I'm going to prevent what I call a public health frame for this response to the pandemic. And as we think about how we might do this, it really thinking about grounding our approach in the ethics of public health is critical because the COVID pandemic is indeed a public health emergency. And Larry Gostin really nicely summarized this, the goals of public health when he wrote, twin moral impulses animate public health, to advance human well-being by improving health and to do so particularly by focusing on the needs of the most disadvantaged. And what you see when you look at this definition is there's this focus on promoting not only population health outcomes, but also doing so in a way that promotes equity and fairness. And so what I'd like to do to close is to prevent present four strategies that governments and health systems could adopt that would allow them to simultaneously save lives and promote equity and triage. And these are ideas that Bernard Lowe and I wrote up in a paper that was recently published in the American Journal of Respiratory and Critical Care Medicine. And what I want to really flag for you is that this article presents an updated triage framework compared to the one we published in JAMA last year. And these updates are really largely around adding more strategies to promote equity while we pursue the goal of saving as many lives as possible. So for your reference, this is a table from the manuscript in which we summarize nine strategies to that we propose to mitigate inequities in the paper. And as I mentioned, I won't focus in this talk on this middle group, the procedural justice considerations around triage teams and discharging their responsibilities, but I refer you to the paper to read more about them. All right, so strategy number one is promoting equity. The first strategy to promote equity is to prevent the need to ration ICU care by ensuring that robust inter-hospital transfer mechanisms are used. And Arizona, the state, the U.S. state of Arizona is really an excellent case study of why this step is so important. There was a very large surge of cases in the Navajo Nation that is in Arizona. This is the northern area of the state where there are very few hospitals, in part due to chronic underfunding of the Indian health system. And these hospitals weren't able to independently handle the surge in demand during one of the early spikes in COVID cases. But at the same time, simultaneously in other areas of Arizona, like Phoenix and Tucson, there were empty beds in the better resourced hospitals. And so if there was really no effort to load balance, this would result in a situation in which Native Americans would bear the burden of disproportionate deaths due to the pandemic. And again, arising from social determinants of health. However, early in the pandemic, the Arizona Department of Health and Human Services created what's called the Arizona surge line to address this. And this is a cooperative system among nearly all of the 130 plus hospitals in the state to both coordinate and facilitate transfer of critically ill COVID patients in order to load balance across the state. And the result was that in 2020, there were more than 5,000 patient transfers of COVID patients from one hospital to another in order to ensure that they did not need to resort to triage, which ultimately they didn't or have not so far, despite Arizona being one of the hardest hit regions of the United States. All right, so that's number one. Number two is exclude triage criteria that would disadvantage already disadvantaged groups. And so there are several parts of this. The first is to exclude criteria that would result in invidious discrimination, such as gender, race, sexual orientation, perceived quality of life or disability status. And then the other part is to exclude considerations that would exacerbate disparities if implemented. And as I've said before, long term life expectancy is one of these. So I think a really important part of triage frameworks is not using long term life expectancy as a triage criterion. Our third recommendation is to apply a correction factor to triage scores to reduce the impact of structural inequities. And I think it's really, really important to note the purpose of this would not be to put any group ahead of others, but instead to create an equitable starting point for a save the most lives approach to triage. And one way to do this is to use a metric called the area deprivation index to adjust each individual's triage score, particularly for individuals who are likely to have experienced inequities in social determinants of health by virtue of living in socioeconomically disadvantaged neighborhoods. This is strongly, the deprivation index is strongly associated with higher COVID death rates. And it's also the case that individuals who are persons of color disproportionately reside in high deprivation areas. So for example, one strategy might be to subtract points from individual triage scores if they're from a highly disadvantaged community. And I'll show one way to do this on a subsequent slide. And so this is just a screenshot from the website to calculate the area deprivation index. It's grounded in Health Research and Services Administration data in the United States and made usable by Amy Kind and colleagues at the University of Wisconsin. Basically, you enter a patient's home address and it provides a rating of the degree to which or the degree of disadvantage of that neighborhood. And this picture basically shows an area of Pittsburgh where I'm from, where a wealthy neighborhood called Shadyside abuts an economically depressed neighborhood called East Liberty. And what you see here is that in red are the economically depressed high ADI areas, and they butt right up against very wealthy areas. And in fact, that line you see is Penn Avenue and Center Avenue. And that is, in fact, a geographic marker that really separates the wealthy from the poorer neighborhoods. And then the fourth recommendation is to give heightened priority to frontline essential workers. And this is not only doctors and nurses who are obviously frontline workers during the pandemic, but also the diverse group of healthcare workers who experience heightened risk of contracting COVID. The maintenance staff in the hospital, the transport teams that move patients back and forth, as well as food services workers. And it's not only healthcare workers who are frontline essential workers during the pandemic, it's also individuals who cannot work remotely, and yet their role, their job, is crucial to the public health response. And we see this in bus drivers and food services workers, grocery store clerks and pharmacy staff, all individuals designated by states and the federal government as essential workers during the pandemic. And I think it's really important to note the justification here is it for priority to these individuals is not that these individuals have more intrinsic worth, but instead it's because of their instrumental value in safeguarding the public during the pandemic. And this is also called the multiplier effect. Another way to justify giving heightened priority to frontline workers is what's called reciprocity. We owe them some obligation for putting themselves at risk to help others. All right. So how can we simultaneously attend to all of these different considerations that I just raised in ICU triage? And there is a feasible solution, which is called a multi-principle allocation framework. I'll show you an example on the next slide, but the idea and the precedent really comes from how we allocate scarce organs for transplantation. There are many factors that go into the point system to determine priority for organ transplantation. We've also used a very similar framework to allocate remdesivir at the time that it was scarce in Pittsburgh as part of a weighted lottery that incorporated multiple considerations. And so this is one way to put all of these criteria together. You see on the top, the top two criteria are designed to promote population health outcomes. The main score driving criterion on this eight point system is prognosis for hospital survival, which could use the SOFA score or another validated instrument. Second, we give more points to individuals who are expected to die within a year. And again, more points translates to lower priority, but we make no other distinctions based on long-term life expectancy. And then there are two criteria that are really designed to promote justice and equity. The first, as I said, is subtracting a point for individuals who reside in highly economically disadvantaged areas translated to an ADI score of eight, nine, or 10. And then second, subtracting a point from the total triage score for individuals who are essential workers in a high-risk occupation. All right, so I'm going to end here and say that if we fail to take steps to mitigate the effects of health inequities on COVID-19 outcomes, at the end of the pandemic, racial minorities and the poor will bear a disproportionate share of deaths, which raises quite serious concerns about fairness. And that there are, in fact, feasible ways to simultaneously save lives and mitigate inequities with ICU triage. Hi, my name is Brenda Punn, and I am the Director of Data Quality at the Critical Illness, Brain Dysfunction, and Survivorship Center at Vanderbilt University. And today, I'm going to be talking with you about patient and family engagement in the midst of a crisis, and really thinking about various ways that we can support families and patients during this crazy season we find ourselves in. I have nothing to disclose for this presentation, and we're just going to dive right in. So when we think about family-centered care, we want to think about what are the benefits that we know of family-centered care. And when I use the term family today, it's really a broad definition. It's anyone the patient considers to be family. So that's family, friends, whatever significant others in their life really falls into this category. And we know that when family are present and engaged in a patient's care, that that prepares them for decision making that they often, for our critically ill patients, will need to make. Additionally, they're an advocate for the patient's voice and their wishes and desires. They are our representative into this patient's life when the patient cannot speak for themselves. And having them involved really helps mitigate the patient's critical illness experience, because once they go home, this is the person or groups of persons that will be processing this experience with them. And also, it prepares the family for making really difficult decisions and the caregiving demands that will happen at home. So it's really important for a variety of reasons for both the family unit and the patient. And as part of our standard bundle of critical care in the ICQ, the ABCDEF bundle, family is a core component of that. Family engagement and empowerment is really an essential piece of the everyday practice that we should be using in our units. And what I mean by this, of what in normal times we'll call it, that we use here is thinking about family presence, family engagement, and family empowerment. And by family presence, the things that we do to make sure that they can be there is have open visitation hours and presence during procedures, even CPR, having redesigned our space so that family has a place in that space to sit and to be comfortable and to even spend the night if necessary. Family engagement is really a step above this, that once they're there with us, we make sure they're oriented to the unit, that they know what we normally do, what different machines in the room are. We provide education and suggestions to them on ways that they might be involved in the patient's care, everything from holding their hand to lotion and range of motion to assisting the patient with feedings and different things like that. We begin to teach them what they're going to be doing when they go home with this loved one, and then we provide regular updates and discussions on the patient's care and prognosis. A step above this is empowerment, and what I mean by that is it really provides the patient and the family a seat at the table where they're actively invited to patient rounds and they become active participants in the shared decision-making that we have. And so this has all been challenged because of COVID-19 in many ways in our units across the world, and there's extreme isolation that's occurred because of various things that we've done to limit the spread of COVID and also to maintain some sanity in our really busy over full capacity ICUs. We've restricted visitation in many of our hospitals and completely limited it in some places. Even when there are visitation allotments, families haven't been coming because they are afraid of contracting the disease. They may not be able to come because they're under quarantine either because they have the disease or because they have been exposed. There's reduced communication. It's really strained with how to reach the families, maintaining contact with them, and having the time to do so. And then for the patient, there's limited interaction or reduced interaction with the staff, especially the COVID-positive patients, and various times to protect our PPE, to limit the number of people that are exposed to the patient. We've done a variety of bundling of our care as we go into the room, and all of these things have had an impact. They've impacted the family, causing fear, uncertainty, anxiety. They've impacted our decision making and their ability to make decisions. When we ask them for consent for various procedures, they're really clueless in some instances as to what is going on. It's an unfamiliar situation to them and really has impaired the discharge readiness for many of our patients and their ability to go home. For the patients, it's caused some of the same reactions, fear, uncertainty, the anxiety. They've lost their voice, in essence, as there's no one there to advocate for them and to be their North Star. Many patients tell us when they're coming out of sedation that when they see their family, they know everything will be okay. And that piece is really missing. And we think that this is also a delirium factory, that these patients are at great risk for delirium and all the negative sequelae that come from that, such as post-intensive care syndrome. On a personal note, my brother was a non-COVID patient that was in the ICU a few months ago. And that ICU, the whole hospital had no visitation at all. He was incredibly ill, multi-system failure, and his wife experienced all of these, and he did as well. And as a family, it was really hard for us to communicate with the staff, to know what was going on, to understand that, to make decisions. Thankfully, he's since recovered, but this was a very trying time for him and our family. When we think about delirium, that last component, one of the things that we at Vanderbilt have been thinking that this pandemic has really caused this perfect storm for delirium. And we've created this mnemonic, this F-COVID mnemonic, of the causes, of potential causes of delirium in these COVID-19 patients. And as you can see, right at the top of that list is family, which means isolation, that these patients, this isolation, we think has really negative impact on the family members. And in fact, it launched us into a study called the COVID-D study. And in this study, we worked with hospitals around the world and recruited, enrolled over 2,000 adult COVID-positive patients from 69 sites that were from 14 different countries. And we wanted to say, what is the prevalence of delirium in this population? And what are the associated risk factors? And we won't dive into a lot of these details that were published in early January. But what I really do want to highlight is that around 90% of these patients at one time or another during their ICU stay were on the ventilator, invasively, mechanically ventilated. And we saw that 80% of the patients were in a coma for a median of 10 days, which is dramatically different. We're used to numbers of maybe 3% of patients are in a coma for three days. And so dramatically lower than this, dramatically lower than 30% of patients in a coma. And then the delirious patients, we saw 55% of the patients experienced delirium for a median of three days. So if we bundle this acute brain dysfunction together, we can see that the patients had coma and delirium for two weeks of their critically ill stay. And you can see on this graph, if we look at the blue lines and the red lines, these dark blue lines and the red lines, bars, that if you look for seven days, that more than 50% of their days are in a, 50% of the patients are in a coma. And then when we take it out to 14 days, you can still see that many patients are having this acute brain dysfunction up to this two week mark, when normally in our really sick, critically ill patients, this is less than a week. So we can see that this is having a real impact. So we wanted to understand what are the, what is the prevalence of the risk factors that are associated with this COVID in these patients and associated with delirium. And you can see that a large percentage of the patients experienced sedation with, especially with benzos, which is pretty different than our usual course for at least a week, a median of a week, that not many of them had awakening and breathing trials. But the thing that I really want to pick out for this presentation is the number of family visitations. And you can see here that in the 20,000 days of data that we have in this study, 17% of them had some type of visitation and only 8% of days had a person, in-person family member present. And we want to know, was that, did that place the patient at greater risk? And what we found was that it did, that the two strongest predictors of delirium was those sedative infusions. The benzodiazepines put the patients at greater risk for delirium. And when family was present, either virtually or in-person, it was protective against delirium. So when we go back to thinking of family-centered care in this time of crisis, and we think of presence and engagement and empowerment, we can really need to be creative because it matters during this season. It really matters. And so we need to deploy the creativity of all of our great staff on how can we get presence, engagement, and empowerment happening here. So I've reached out to multiple sites and asking, what, how have you done this? And looking at ways that we can improve at Vanderbilt and ways that I can share in this presentation. In some ways to get presence is in our COVID units, we're putting a seat outside the door. So their family's not allowed in the room, but we give them a seat outside the door and a time frame to be there. The staff knows they'll be there from one to three. They have a place to be there from one to three every day. And so that facilitates all of the rest of the things on this screen, the engagement and empowerment. In the non-COVID patients, we allow a visitor in normal visiting hours and that one visitor per day. And we've also had to really facilitate virtual visitations. Many of you over the summer has got iPads and phones and support devices, securing Zoom lines and putting those right on the iPads and so that they can be used and deployed. And really using volunteer staff to help be these communicators. Or as staff shifts in the hospital, using the staff from the clinic area or the acute care area to help be these communication facilitators. When we think of engagement and empowerment, I put all these suggestions in one bucket here, that we can provide the same general orientation materials that we usually provide by mailing them or posting them on the web or Google Doc that we can send a link. We can do a phone orientation. We can create one of the things that we've done at Vanderbilt, patterned off of some of our colleagues in Louisiana, is creating a designated communicator. And for us, this is a volunteer physician that joins the team every day in rounds and then at the end of rounds, contacts the family to provide the updates. So this frees up our treatment team to do the bedside things that are needed. And it also provides strong communication to the family and consistent communication. Another unit that I've talked to uses a, created a pop-up Zoom support group for the families. This provided a way that they could do the orientation and describe these are the ways we normally do it in our area. But it also provided opportunities for those family members to see what others are going through and to help one another. And one of the last things that I'll share is really that we need to recruit and equip a team to help us in this way. And we have lots of people asking how they can help, what they can do. Chaplains are huge in this situation, that they can be there to provide support and communication in presence to the patients. They can be that go-between for us, social workers, and again, that volunteer group that we've been, I've already alluded to. Just in summary, as we talk about how to get family present and engaged and empowered during this crisis, is to remember it's important that we see in the COVID-D study that family presence mattered. It reduced the risk of delirium and what we think is all the negative things that would be associated with that delirium. And we need to be creative, ask around, ask in your areas, deploy some of these particular strategies that we've talked about today. And in this crisis, let's go forward and not backward. When you are out of a surge situation and can handle things, deploy new creative strategies so when the next busy season comes, you're ready to face them. Thank you for your time today. And feel free to reach out to me with any questions. I'm happy to help. You can reach me at vanderbilt, brenda.punn, at vumc.org with any questions. Thank you.

critical care

resource allocation

health inequities

triage policies

family-centered care

pandemic restrictions

equity considerations