All right. Good morning, everyone. Testing equipment here. All right. So welcome. This is a session entitled Drinking from the Fire Hose, Leading in an Intensive Care Unit. And this could probably be a week conference or a two-week conference about how to navigate this, but we thought we'd bake in a few interesting areas of focus for folks that have to lead in an ICU. So before we start real quick, how many folks in the audience are ICU leaders, whether it's a medical director or an administrator or something like that? Okay. And mostly medical ICUs? Mixed. Mixed med surgeons. Okay. And how many of you have had formal leadership training as part of your job? One. One. Yeah. And as we were kind of putting together this talk, that was one of the things we were noting and thought, you know, having these sort of topics at NHS would be reasonable for folks to try and digest things. So with that, I'm going to start. I'm Will Bender. I'm one of the pulmonary critical care faculty at Emory. I serve as a medical director of a medical surgical ICU in one of our hospitals, and then also a site director for one of our hospitals. So a number of ICUs. I have nothing to disclose. And so what I'm going to kind of briefly chat about is just, you know, the development of physician leadership and healthcare over time here. Talk briefly about the team-based care approach that we use in the ICU. Look through some of the leadership skills that are necessary to lead an ICU. And then also kind of focus on this idea of leadership style and how I think working with leadership skills can help work, help create an effective ICU team. So the idea of physician leadership and healthcare is really kind of grown over the past 30 years or so. And there's been, you know, seminal events through time to kind of help with that. I think, in particular, as healthcare leaders on the business side of things, executives have realized that physicians, with their knowledge of medical science and patient care, you know, offer a great opportunity to lead. And there's some corollary evidence that suggests healthcare institutions that have physicians, particularly in positions of CEO, actually do a little bit better from a financial standpoint. Obviously, it's corollary data. It's nothing perfect, but it does promote kind of some food for thought with that. And at the same time, with physicians and their, you know, style of training and focused on the patient, it makes, you know, quality and safety efforts to improve a lot easier to initiate because, again, they're coming from a patient-centric focus. Then I think most recently, obviously, something all of us just experienced, the COVID-19 pandemic really kind of put all of our leadership skills into overdrive as we were creating ICUs in the middle of the night, you know, re-energizing, reconfiguring our workforces, both at the physician level as well as, you know, if you're in an academic institution, house staff or APPs or stuff like that. And then even on the education side, as we look at our education pathways, and over here on the right, these are some of the entrustable professional activities for the Pulmonary Critical Care Fellowship. This was published in CHEST a few years ago, but one of the entities that is basically expected of fellows when they graduate from their pulmonary critical care training or critical care training is the ability to lead a multidisciplinary care team. So it's interesting that all of us in the room do some degree of leading, although we acknowledge that we haven't had really any formal training as part of that pathway, although it is clearly stated in what we should be getting when we go through our training pathways. And I think, you know, I like to include this slide in a lot of my talks because, you know, critical care is obviously very team-based, but the focus is entirely on the patient. This is from an article in Critical Care Medicine back in 2020 by Stallings, where they were looking at optimization of rounding practices to improve adherence to the ABDC-EF bundle. But again, it highlights the fact that our team in the ICU in particular is incredibly varied, very big, diverse, but everyone's singular focus is on the patient. And so when we look at, you know, the patient-focused model, the patient-centered model, there still needs to be some semblance of leadership of this team in setting up an environment for folks to excel in when they take care of that patient. And the SCCM Task Force on Models for Critical Care a few years ago noted that an intensivist-led, high-performing multidisciplinary team dedicated to the ICU is an integral part of effective care delivery. And a variety of studies have kind of, you know, reinforced this concept. I think one of the most interesting ones is from 2010, a seminal paper from Kim at, I believe, Pittsburgh, where they looked at a number of ICUs in hospitals in Pennsylvania, looking at some physician staffing models as well as 30-day mortality outcomes. And they basically found a pretty strong relationship with high-intensity physician staffing as well as a multidisciplinary care rounding process. So having that physician-led team with a multidisciplinary group in tow delivered benefit to patients. I think as we pivot now in thinking about, okay, so we know that physicians have an important part in leading the team. What sort of skills are needed for that team? And what sort of skills are needed for the physician to kind of drive the team? And again, this could be a talk in and of, you know, multiple talks in and of itself. And there's multiple models of generic leadership competencies that are out there. The National Center for Healthcare Leadership has about 26 individual competencies that they bundle down into three domains, which they call the transformation domain, the execution domain, and the people domain. I happen to like this table from Dr. Stoller, who's out at the Cleveland Clinic, who does a lot of work with organizational development. And he kind of boiled it down into basically six areas or so. And I think looking at these as competencies that gets your foot in the door, having these competencies in place gives you some street cred as an ICU leader. And, you know, having technical knowledge, technical skills enables folks to basically be able to speak to the nuts and bolts and operation of the team, having overall knowledge of healthcare. And obviously, it's a very broad statement. But I think when we think about an ICU leader that speaks to, you know, the global healthcare environment, but also your local environment, you know, what are your protocols? What are your rules and regulations? How do you handle certain situations? Those sort of things. And then, you know, good problem solving skills. So the ability to resolve, you know, local challenges, local behavioral issues, manage projects, you know, strong communication skills, your ability to lead groups, negotiate. Obviously, with all of us working in ICU leadership, so much of what we do is just conflict resolution and negotiation and bartering and all those sort of things. And then I think a commitment to lifelong learning, as Dr. Stoller highlighted here, is very important as well, particularly in the setting of a rapidly and changing, rapidly changing environment based on recent evidence, guidelines, you know, again, get harkening back to the COVID-19 pandemic over the course of two years with that, how much of our practice has changed with that and being able to implement a lot of the knowledge that we learn. And then lastly, you know, I think ranking all of these, all of us would probably rank these a little bit differently, but I do wonder if emotional intelligence, which he lists on here, is one of the most important ones. You know, the ability to know, to reflect upon oneself, one's abilities, and then put that into context with relationship with other people. I think that might be one of the most important skills that you have to have as you lead an ICU and lead an ICU team. I think it's very likely, I think, that more than skills are needed, however, when you kind of think about all of this, you know, those skills that we just highlighted, those are things that some folks just have, their innate abilities, but at the same time, they're also things that can be coached and taught and learned. But at the end of the day, they're still just skills. You need to be able to apply them in the right context. You need to be able to know when to use them, how to use them, and those sort of things. And I think that's where thinking about leadership style becomes very important as well. You know, a promotion of different ways of utilizing all of those skills to accomplish the goal, and in this case, taking care of a critically ill patient in an appropriate team-based environment. And again, this could turn into a week conference in and of itself, but there's basically been four types of leadership style that have been predominantly associated with healthcare. The first is a transactional leadership style, and I think we're moving away from this in medicine as medicine's model changes a little bit, but it's traditionally been promoted in medicine due to the kind of the medical training structure that we have, the hierarchical concept of medical training. And basically, the ICU attending is seen as being exclusively responsible for care planning and decision making, and the ICU team is there to provide information only. I think this is a model, again, that we've moved away from, particularly on the critical care side, as folks have embraced multidisciplinary rounding and that sort of thing. And again, I don't think this is ideal because it runs counter to the idea of multidisciplinary care. There's the adaptive leadership style, which is used to enable a group to overcome challenges created by change. I think the challenge with this model is that it can create a high level of distress, given that it's very, very blunt. The next model, I think, that does get a little bit more street cred at this point in time in healthcare, and particularly with critical care leadership, is this transformational leadership style. Basically, utilizing large-scale inspiration for a mission. However, this tends to kind of really steamroll right over the idea of patient autonomy and letting patients kind of help dictate their care because the patients are subjected to the inspiration or the focus imparted upon them by the ICU leader. I think the last model is probably the most appropriate and realistic for the majority of things we handle as ICU leaders, and that's the servant leadership model. And with that, we look to build a community in which team members are committed to putting the patient's interests first and lead team members to achieve the goal of providing high-value patient care. Now, obviously, it's not this concept of creating an environment where we're kind of lifting the entire team up together, encouraging everyone to be part of the patient's care plan is not always going to fit in every situation. You know, a group discussion during a cardiac arrest is probably not the ideal thing for the patient. But I think the idea of establishing psychological safety as part of this style enables those more emergent situations to be handled in a smooth and more effective manner when you've created that level of culture. And I think the last thing I just kind of wanted to highlight here is when we talk about this concept of leading, particularly with the servant leadership model, we're looking to produce change as leaders. We're not necessarily managing. With management, our aim is predictable. We're looking at orderly results, you know, basically taking the, you know, taking the, you know, the act within a culture and working on that, whereas leading, we're looking to produce new culture or a new focus with things. So in summary, I think physician leadership and healthcare has continued to grow for a number of reasons, and ICU physician leaders definitely need a number of important skills to be effective. With that being said, I think leadership style, particularly the servant leader model, I think likely plays more of an important role in the development of effective ICU leaders than just a running laundry list of skill sets. Thank you. Are you stuffy in here, huh? Yeah. Cool. My name is Josh Dinson. Thank you guys for showing up at 715. I'm trying to try not to talk to you. I know half of these people in the audience, which is always exciting. I'm assistant professor of medicine at Tulane. I'm director of medical ICU there, and wear a lot of other hats, and I have nothing to disclose. So objectives, I wanted to, I'm tasked to talk about managing ICU strain, and I want to talk about how it's defined and recognized. There's actually a lot of literature out there. Many may or may not be aware, but I learned a lot about it. I'm by no means an expert in this, but I think all of us deal with this a lot. We'll review the consequences of it, and then I want to kind of put out some strategies that you probably already use and other ones that you may not be using to try to manage it. Not all patient volume is bad, of course. Strain is defined differently, and I'll talk about that briefly next, but many outcomes are better with the more volume you have. You know, I work with a lot of trainees in our ICU, some are here, and sometimes we talk about, well, how can we cut down their census? How can we cut down their workload? It's not necessarily related, because if you have too low of a workload, then you have an increase in mortality. At least in ICU, this is known in angiography, for cardiology, and other situations, but in the ICU, this is from a 2006 New England Journal paper with Jeremy Kahn, and showed that mortality improves with more mechanical ventilation in your ICU, but that's not what we're talking about today. We're talking more about when this becomes overburdened, and we all just went through COVID, and there's no need to really rehash that. Obviously, every ICU was strained in that circumstance at one time or another, but depending on how you look at it, it could be defined differently. Is it due to patient volume, just an average census? Is it due to, I got 20 admissions yesterday, like am I strained now? Or is it due to the staffing, like there's not enough doctors, there's not enough nurses? Physical bed space is another thing to consider. We all probably see people in the ER sometimes, or at least my hospital does. Does anyone else's hospital still get patients stuck in the ER? All right, other ones are working more functional than mine, I guess. Patient acuity is another thing. One sick patient will take up your entire day or night, as you know. The other thing that was really interesting in this paper was, you know, Scott Palper in Pennsylvania has a lot of literature on this. There's a number of recurring names you'll see, and this looked at the perceptions of providers, nurses, and physicians, and kind of how they perceive strain, and they perceive them differently. So physicians pretty much viewed it most strongly as patient census-related admission criteria and severity, while the nurses actually took a more holistic approach, not surprising. They thought more about bed space. They thought more about staffing, because I feel like they're always worrying about staffing, and I'm not, right? I don't know if you all have that experience or not. I'm like, well, just let me know what you need me to do. But that's something to take into account if you're leading an ICU, that people view it differently. The consequences, I don't have time to review all the literature, but I'll hit some main teaching points that I thought were really relevant from the literature. Triage decisions, ICU strain, depending on how—did I pass the definition sign? No, yeah, there's a definition, which I'm not going to read. You guys can tell what it is. When the ICU is under strain, whether it's from—usually the most common definitions relate to volume, census, and acuity. Triage decisions move quicker, so people move through—you guys already know you do this. The literature shows they're readmitted more often, but the outcomes do not necessarily change. However, in general, this is a little bit mixed, that mortality is increased when the ICU is under capacity strain. COVID is a good example of this, which I think the literature is more clear there. A lot of what I'm talking about was pre-COVID, and what was interesting in the studies that have been done to date, this finding was most prevalent in closed ICUs, where your resources are a little bit more finite, and you have to cover all those patients, versus an open ICU, where, okay, the hospitals can help you on a few other patients, et cetera. And they've looked at other studies related to closed ICUs specifically, and this is not just in the U.S., it's also in the U.K. and other places. Shorter time to DNR, which I don't know how I feel about that. I mean, I notice this. I'm sure some people do, depending on the practitioner or the hospital you're at. I mean, goals of care discussions, you have less time, and sometimes the decisions are just made quicker. It leads to shorter time to death, but when they looked into the granularity of the data, it looked like the DNR orders were just coming in quicker when the ICUs were under strain. Certain diagnoses are more at risk, sepsis, respiratory failure. They're less likely. You guys all do this. I do this. If the ICU's busy, we're going to triage a little heavier in the ICU, versus not, depending on, at least I do this in our ICU. And the outcomes show that those people usually do worse on the floor, but again, you have to kind of make some cutoff somewhere, and you need help managing the patients occasionally. This graph on the right was from a paper that looked at respiratory failure, and on the y-axis, you have a predicted probability of ICU admission, and the bottom, they use a strain index. So as their strain increases, the probability of being admitted to the ICU decreases, and it's variable across different hospitals, which is another big point of the heterogeneity there. Less time spent in a new ICU admissions, duh, and less likely to receive dermal prophylaxis. So this is just a snapshot of the main articles from this, the consequences of strain, but you could really get in there. I mean, there's more that I haven't really touched on even. It's really impressive. Briefly on intensivist patient ratios, the one on the right was from Haley Gershengorn, and it was a data, I believe, from the UK, and kind of you guys have probably all heard about this. What's your average patient census in an ICU per intensivist? The data is a little bit mixed here. It's not clear-cut that this is how many you should have. In general, on the right, the UK data, the average mean they had was 7.5, and I found this surprising, because in the U.S., I feel like most of us are, we always think about the ideal is like 14. I don't know, that's what I'm always told. Anyone else have numbers that they think of? What's your ideal? Anyone? 10 to 12. I think that's what we all think, and that's the left side is the U.S. data. This is from 2023, and the average, I believe, the median was 11.8 around 12. So just in general, like how many patients should you be covering? The right side showed a u-shaped aspect where more patients had to higher mortality and too few patients at higher mortality, but it wasn't like so clear-cut that this is how it was. It's just, look at the variability across the different ICUs. Different people just do different things, and it depends on your unique characteristics, who you're caring for, and how sick they are, all the things we talked about in the beginning that leads to strain. It'd be silly not to talk about burnout and strain. I'm biased because I do some research on burnout, and so I want to talk a little about that, but I think that they're just so intertwined. Obviously, the pandemic, we all saw this, but even after the pandemic, I think people are waking up. Before other news that happened the past three days, we had a lot of news talking about the nurses and the health care workers going on strike. I mean, in this year alone, we've had two major health care centers in different sides of the country on go on strike, and a lot of it has got to do with this, with strain, with burnout, with just quality of life in general, I think. It's not a problem strictly in the U.S. It's across the world. How many guys, how many people are practicing outside the U.S. in this room? Anyone? Okay, just wondering. This is what led to strain. When you ask people, all right, what causes your strain? What causes, what makes you feel this way? Inability to discharge was one of the top reasons. Increased patient volume, increased acuity, inappropriate admissions. There's a theme as you read these. There's a theme of like using, not using your expertise appropriately, and it's not just physicians. It's also nurses, and it leads to more stress. It leads to more feelings of distress. All of these are basically 90% of people said this is what's causing this, and I think it's worth reviewing these perceptions, but I want to keep moving too. This was a study that we were part of that was led actually out of Emory with Dr. Agarwal, and it was, it's really great, like, and I was part of it, and we were a site and all this, but I was digging into the paper, and there's some really good data, things to look in there if you guys are leading an ICU that just talks about, like, what are academic, at least academic ICUs. This doesn't approach community ICUs, but what are, what's everybody doing around the country as of last year or two years ago, and the structure of the ICUs, and how people are practicing. I think it's worth reviewing. In this paper, one of the big take-home points was the ICU strain is associated with burnout in these higher workload ICUs, and there's a great figure, a DAG, and how workload relates to burnout and the different factors, so the ICU structure factors, the patient-related factors, the physician factors. Just, I would, I think it's a great paper to read if it's in critical care medicine, if you are leading an ICU, just to say, hey, what are other people doing? How many patients are they managing? What is their census like? This has a lot of that data in there. It's really interesting, and can workload be reduced? So, how can we manage this? This is a, from that other perception study, and that study actually was out of Alberta, Canada, and it took the entire province, is it a province there? I'm not sure, province, and took the entire province into consideration, and you can look at what people, again, what people thought would reduce their level of workload and burnout. Again, more beds, goals of care discussions, nurse-to-patient ratios, and nurse-to-patient ratios, contrary to intensivist patient ratios, actually has pretty good data. Sorry, y'all were taking a picture. I was going to let you take it if you want. The nurse-to-patient ratios has pretty good data to support that if you have a mismatched ratio, then it definitely leads to worse outcomes, and this graph on the figure kind of just was from one of the big papers that showed that the number of nurses per bed, as it increases, mortality goes down, and it seems to benefit sicker patients the most. Again, not surprising. I probably have, what, a couple minutes, sir? This is a paper that, this is a study I'm kind of interested in. Well, obviously, we did it at Tulane, and it was like, basically, I'll kind of get into it. So, when I started in New Orleans, there were some med psych residents doing these things called death cafes. Has anyone ever heard of this? So, they adapted this community group discussion therapy type thing that happens out in the community around the world, and I think it started in England, where people get together and they talk about death and grieving and things like that, and it's meant to help them with coping strategies. So, these med psych residents had kind of adapted it to the healthcare environment and started doing it at one of the hospitals we work at, and just getting people together from all around the hospital and talk about anything, but related, it's supposed to be around death, and you're supposed to have cake at the end of it. That's, like, the only rules, and you can talk about whatever you want. And so, I went to one, and I was like, oh, this is pretty cool. You should, like, study this. So, with one of the chief residents, who's now a fellow at Brigham, and me and all trainees, basically, designed this randomized controlled trial to randomize people to a death cafe intervention that worked in an ICU, and the inclusion criteria were you had to do one work week in an ICU in the past four weeks, and you could be a nurse, a therapist, a pharmacist, whatever, and we did it multi-center around New Orleans area and basically measured outcomes. Now, I don't have all the data to show you because it hasn't been published yet, but I did want to put up just kind of some of the qualitative data. We had a qualitative analysis done with a patient scientist I work with at Tulane, and these are the themes that people reported that they cope with. So, I thought it's relevant for you guys as ICU leaders to see what are people reporting in, you know, with anonymity. What are they doing to help themselves cope? And it's, like, simple things. It's exercise. It's sleep. It's time with family and friends. It's all the things that we already do that sometimes when you put it on paper in writing or tell people to do these things and prioritize, it can kind of help. And then this is just some of the quotes of what their experiences from the death cafe were, and I thought it was worthwhile because most people felt it to be a positive experience, a safe place to share thoughts. A lot of people thought it was a non-judgmental environment, and so we got people together. I don't know if I said this or not. Originally, it was all through the pandemic and supposed to be in person, but we did it at Zoom, so it has its complications, but you get together with anywhere from two to ten people. It's ideally be kind of a small group, but they just kind of made it, and it's led by a psychiatrist, but it's pretty practical to do, and in general, people found it to be valuable. Again, I don't think adding something extra fixes burnout in the ICU. We don't have a fix for burnout in the ICU or anywhere, really, but it is something that is an option that some of you might find useful because we've gotten a lot of emails just from the protocol paper about it. If you have questions, let me know, and so take-home points. I see strain are perceived differently. The literature is pretty robust. Obviously, it's associated with negative consequences. A lot of this stuff, managing it, is out of our control. I don't have a lot of money. I can say, all right, I'm going to hire 10 more intensivists to come and help with everybody. That would be great, but there are some things we can do that are in our control, and that's what I'd focus on. Thank you, guys. Good morning, everyone. My name is Miriam Kaus. I'm going to pull up this some slides. This is supposed to be a panel discussion, so on my talk, I'm going to be discussing leadership principles to help drive quality improvement. When you think about quality improvement, I know a lot of the times, at least in our department, we focus on this a lot, but for the most part, at least in most of the faculty, we just see it honestly as an added burden unless you think that there's something really interesting to you. What I want to go through is some of the principles that I think we can try to utilize to come up with creative ideas to come up with good quality improvement projects. I'm at UT Houston. I'm one of the assistant professors in pulmonary critical care medicine. This is my child. You need teamwork to stand up, so now he's running around, but he needed that initially, so important in quality improvement. Then I also am one of the cardiac critical care unit medical directors, and I'm from the pulmonary critical care side. Just like my colleagues here spoke about, we didn't actually get any formal leadership training, but oftentimes, especially when you're in academics and you're trying to build a name for yourself, if you demonstrate some leadership qualities, you get put into these roles, and then you have to figure it out on your own. I'm going to review a few of the leadership principles as I discuss, and then I'm also going to discuss theories actually from other disciplines that may be successful in medical leadership. So a little bit less database. So again, first and foremost, teamwork is going to be key. So we have to realize that the hospital system is huge, and we cannot do this alone, especially in something like quality improvement. I think when we think of coming up with projects, we don't think about the nitty gritty of who is going to do the data collection and all of that, unless you have a very robust system already set up in your hospital. So teamwork is super important. And in the ICU, we're utilizing many members of the team, and we really need to learn how to use them appropriately. And then I'm going to go through four topics first, where I think we can use these models that are kind of from a business standpoint to see how to choose projects to prioritize. The first one's going to be the interventors or 80-20 rule. So what does that mean? If you look historically, it's been shown that about 80% of effective change is actually implemented by 20% of the population. Similarly, maybe 80% of the things you see are implemented by people who are the most motivated. So in 1951, one of the business consultants noted that in any population, it's actually a very vital few that contribute to the bulk of the effective improvement. So really, we shouldn't focus on every single thing in the ICU, but rather, as managers and leaders, we should focus on the small fraction that's going to yield the greatest value. So that may be cost containment, mortality benefits, so small things that may lead to the most change. So don't look at small changes that's going to lead to a small change that's going to only give you a very small bang for your buck. The next one is benchmark. And I think most of us do this anyway in our institutions, is to compare ourselves to other institutions that are similar in size. This is a little bit harder to do in health care, because every hospital system is so different. One of the ways, at least in the ICU compared to the wards, where it's a little bit simpler, is to use something like an Apache score to see that if you're comparing the similar patient population compared to other hospital systems. So defining quality in health care is not as easy as defining the outcomes in business, where usually those are monetary outcomes. So in the ICU, we often use mortality rates. But again, that can also vary depending on your patient population. So just something important to remember. The next thing that I think most of us have been a part of is a root cause analysis. So when I think back to root cause analysis, at least when I was a trainee, I always thought, you know, again, I think most, is anyone in the surgical field here? No. So whenever we think of root cause analysis, or M&Ms, at least in the surgery department, you would think that they're very, very scary. And sometimes even in medicine, I think when we look at the root cause, you almost start to feel like it's kind of like looking for the blame where it went wrong. So even the data shows that the RCAs are actually not super beneficial. And when we look back, Percarpio actually looked at this. And they performed a systematic review of articles related to RCAs and found only 11 that actually discussed any sort of impact of the RCA process. And they saw that three actually reported on clinically relevant outcomes. And I think we can probably all look back and we think of RCAs. Most of the time, we don't lead to a systematic change. Rather, we focus on a very small cause that may have led to that effect. And most of the times, I think RCAs varies really significantly across sites. And they may focus too much on finding a single root cause. But this is one of the things that we tend to use whenever we're looking for quality improvement projects. Instead, maybe a better one to focus on is something called a failure mode and effects analysis. So this is more proactive than retroactive. It does require more thought process from us. So here, you would want to identify a process that you think may have many steps. And you want to identify each one of those steps and then see where the problems may arise and how important they may be. I'm going to show this graph instead just so it kind of gives us an example of what that actually means. So this maybe looks super simple, right? We see patients who are septic all the time. So you identify the patient as septic. So you identify all the steps for identifying that and then identifying your source of sepsis. So you would create this map and all the steps necessary to identify those processes. And then you would do a hazard analysis in the last step, OK? And when you're looking at the hazard analysis, within the hazard analysis, you're also going to look at how are you going to prioritize this. So what is the severity of an issue occurring? What is the probability of it occurring? And then what's the rate that you can actually detect it? So if the severity is low, probability is low, and detectability is low, it may not be worthwhile focusing on that particular step. So as you can imagine, it requires a lot of thought process. But this may be a lot better than utilizing something like an RCA where you don't identify these things, right? So just something to think about. So that's choosing your quality improvement project. How are you going to design it? So I'm going to just talk about metric identification and data collection. This is, again, it's super meaningful and often difficult. So the steps that I just mentioned can be helpful. But oftentimes, we think about mortality, which is frequently used in the ICU. But it has its limitation. So after you identify your metric, it could be other things as well, right? So how you're choosing a certain antibiotic, or you can look at CMS data to identify the metrics, or look at national data to really decide. I had to discuss this, actually, with our chief medical officer because she loves quality improvement. She said that they often look at CMS data and what nationally is going on. And that helps you decide what you need to focus on every year, right? And then we can use for data collection. Now, electronic health records make data collection easier. But extraction still remains a major challenge. So I think instilling help from other team members who are part of the quality improvement project will help you be more efficient. Sometimes, this task is tasked to the physicians. And to be honest, for me, I think that that ends up taking away from our thought process and focuses on the minutiae that's not helpful. And again, this kind of goes back to looking at the effect analysis is you want to think of the frequency of the process and how to proceed. So again, you're going to look at what is high risk, what's a high frequency process, and then you evaluate that by outcome. Whereas if something is low risk, but it happens frequently, maybe you can use other measures, some process measures that are not super stressful. And then this kind of, again, shows an algorithm for appropriate metric identification. So three different things that you can kind of look at, outcome, process, or structure. So each of those, depending on what you decide to choose. For example, if you choose an outcome measure, you're deciding the percent of time antibiotic is given as desired. And then you're going to see, is it identifiable? Are you able to measure it? Is it relevant to your institution? Can you do it in a timely manner? If the answer is yes, then you measured the metric. And when you're asking these questions, it really makes you think. And if you can't, then you go back and decide another outcome that you want to focus on. And then you just kind of follow the steps and choose a project. Next is how you're going to do your data presentation. So this is going to vary between institutions. So maybe at the end of this talk, you guys can kind of also mention how you guys show the data once you actually use it. For us, in our institution, we tend to do either monthly meetings. We use Vizient in our institution. And that way, it really shows us in real time how change is progressing. We look at certain DRG codes, which tells us how our mortality is compared to the rest of the hospital. So that's one of the ways we do it. And you want to choose a little bit of a low-stress environment where you can do it. So data shows that actually public displays of data may help improve compliance. But as you can imagine, sometimes people also take this as an attack. So you really want to make sure that your venue is a safe environment when you do the data presentation. And then you want to draw attention to the items that really have the most impact. So again, going back to Vizient, when we were looking at this data, there's like a list of like 50 things that you want to compare if it was missing the chart, was it coded or not. So within that, I think really we want to focus on the meat of what's going to change your mortality ratio. So you really want to draw attention to items that have the most impact. And then how can we implement quality improvement measures? One of the things we often tend to do in the ICU is checklists. And I think there is data that the checklist can be useful. But it's honestly a limited amount of evidence, right? It's just the best tool that we have. And we tend to use it. But we really want to avoid using too many checklists because we all know that also adds to burnout and adds to a time that we don't necessarily have in the ICU. But once you utilize a checklist that has kind of formulated your rounds or whatever, then you can continue to use those and not continue to create multiple, multiple checklists because I don't think that's useful. And then once you've decided on a measure, you're having good changes, how are you going to sustain that change? You really want to set priorities and buy in from the team. And this is where you need incentives for your team members. So really, what motivates all of us, right? When you look at data or you ask yourself what is going to motivate you to sustain change or lead to quality improvement, three big things, I think. So one is autonomy, something that we just want to do. We want to be better at quality improvement. We want to lead to a change. And so we want to get the support from the administration. And then we want them to get out of our way and just let us do it. The second would be we just want to get really good at something. And that's why we chose this particular quality improvement. And you want to be the best at it. Or three, it's your self-purpose. You're either working for free for a higher goal or it's profit-driven alone. And you would think that maybe the profit-driven would have great results, but it actually has some mixed results. So when you're trying to get buy-in from your team, monetary incentives have some role, but may not have the only role. So it's just something to think about. And I'd love to hear everyone's talks. So take-home points are that extrapolating quality improvement projects from business may actually be applicable and successful in the ICU. Quality improvement really requires buy-in from multiple levels and departments. And then choosing the right metrics and processes, honestly, are just as important as the data collection and interpretation. And we all know that that is probably the most challenging part to the whole quality improvement process. Thank you. Okay. Okay, so I'm going to talk about advanced practice providers or APPs. Like Dr. Bender, I'm from Emory. Here's my objectives, and I have no disclosures. And so just before we start, just a show of hands, how many folks work in a system that does not have APPs? So very few. So it's a pretty common occurrence these days. And, you know, the landscape has really changed a lot. They're now an important force, labor force, in critical care medicine. These are physician assistants and nurse practitioners. And they're really the primary interface between the care team and the patients. And before we talk about integration, we should talk about what APPs can do and what APPs do do. And the APP is really, it's a, the role is complex. It's autonomous. They can do daily rounding. They can do nighttime coverage. They can do procedures. They may or may not be guided by this principle of standard work. And I'll get into that in a second because I think it's important. What they do do in a hospital or a health system really kind of depends on the needs of the health system and the culture of the health system. There's this debate about whether or not we need APPs in this role. I firmly lie on the yes side of things. But those folks that think that APPs are not as valuable as I think often cite the issue that there's a gap between medical knowledge and the practical skills that APPs have coming out of training, which can affect the quality of care and patient satisfaction. And then there's this idea of scope creep. Does anyone know what that is? Right. It's the argument that the APPs are doing more and more of my work, right. They're sort of putting me out of business. So that's a, I think that's a bit of a fallacy of an argument, but it's definitely an argument that's made. Folks that do believe in the value of the APP as a concept like myself really can point to some evidence that APPs actually can improve patient outcomes. And this is bolstered by this large body of evidence that really talks about the fact that we don't need to have 24-7 intensivist coverage anymore. And this was this New England article. I'm not sure if, there's some of you in the audience that look like that you were practicing like I was when this came out in 2013. And I remember where I was. I was in a health system in Northern Virginia, and this came out. And this is basically the argument that outcomes are not worse when you don't have 24-7 intensivist coverage in the ICU. And we did at the time, and that hospital system still does. And I remember thinking this is going to change. We don't need to work here 24-7, and it didn't. But this argument here is, I think, supports the idea of sort of having a more diverse workforce that can sort of cover these really challenging aspects of care. Importantly, this study did not have as the arm of the non-in-house intensivist, this was not exclusively APPs. This was house staff and others. This study actually predated that. This is an abstract that came out two years before that through the American Academy of Physician Assistants. And this actually got more to the point, which is, this was two ICUs, this was retrospective. And this compared an APP-run ICU and a house staff-run ICU. And outcomes were also no different. This said, not all APPs are the same, but maybe they should be. Education for APPs is quite variable. This can range anywhere from a year and a half to five years. And there's a lot of differences in accreditation and requirements. Within critical care, I think we all sort of tend to think of NPs and PAs as the same. But this is not the case outside of critical care, where some nurse practitioners don't even do an ICU rotation or don't have to do an ICU rotation. The idea that APPs should be the same in critical care, though, I think is important. And so a lot of systems are now advocating for APP critical care fellowships. And this is an important thing because, you know, it costs a lot of money to replace or to train an APP. The idea of fellowship is, I think, an important one. And there could be standardized training through an accredited body. This is not standard across the country yet. This is still hypothetical. And so a lot of these fellowships are sort of making it up at this point. That said, I think they are migrating to this standardized training curriculum with some sort of overarching accreditation body. And this idea of standard work, and here's a great example. For those of you that are MDs in the room, we all went to medical school. We all learned how to do an HMP, right? We all know chief complaint, history of present illness, you know, past medical history, right? We all do it the same. So if I left my institution and went to your institution, I could pick up and start working the next day. This is standard work. This is a detailed definition of best practices that's done sort of everywhere. This is not the case in APP land. The training is not, they don't embrace standard work. There's a lot of different ways that APPs are taught how to do things. And I think this is to the detriment of the field. And I think that this is an important part of fellowship that needs to happen. There's equivocal evidence about APP fellowships. A lot of it is sort of self-reported confidence surveys, which does show that the APPs have more confidence when they are doing fellowship. The one objective data point that exists actually came out of our institution at Emory, which was very small. It looked at a couple dozen fellow trained APPs and non-fellow trained APPs. And it basically looked at billing efficiency and productivity and did in fact find that the fellow trained APPs were more productive and efficient at billing than those that were not fellow trained. While this was statistically significant, this was not clinically significant. They calculated that it was around a $5,000 benefit in terms of that billing practice. And that of course, compared to $100,000 to train the APP. So let's get into APP integration. And I think really this sort of needs to be centered around some topics that are a little bit of a third rail, which is kind of redundant work, right? This varies from state to state, but essentially an APP can do critical care without an attending. They can write a note, they can bill critical care time, and that's legal in I think all 50 states. But an APP cannot admit a patient. There needs to be an attending of record for an admission in this country. And so there's this sort of dynamic tension. Should an APP own their own patient or are they really, to use a word that's not overly liked, are they extenders? Are they physician extenders? And again, there are cultural differences within a hospital or a health system as to what the APP should do. And this is not data, this is not research, this is my concept of how these cultural differences exist. And there's kind of a high revenue approach, which is the APP and the attending both see a patient, they both bill critical care time, they bill the 9-1 and the 9-2, and the critical care time mounts up and they kind of collaborate on the care of the patient. And then there's this low cost approach, which is that the APP and the attending see their own patients. There's really a division of labor and subsequent reduced staffing needs. And so I drew this for you in a nice picture. These are four patients, the high revenue approach where the attending in red and the APP in blue each see, each patient, they all round, they all write notes, they all do procedures. I compare and contrast that with the model on the right, which is that each provider sees their own patients. You can imagine that the critical care time runs up a little bit on the left model and you can also imagine that the cost of labor is down on that right-sided model. And to dovetail onto Josh's point earlier, this has implications for staffing ratios. So the SCCM did put out a statement talking about physician ratios. And by the way, the 14 comes from there because they told us. That's why we do 14, right? Because someone told us we had to. So this is the statement on physician ratios. But they mention that APPs are endorsed, but they don't mention the APP ratios. And so, again, you can imagine, you know, if I have a 20-bed ICU and I'm doing that high revenue approach, the physician ratios, there's implications there. Ruth Kleinfeld, an old, a past president of SCCM, did some research on APP ratios and found that current practices are around one to five. This is not data or outcomes driven. This is just practice, right? So this is just a survey. And likewise, on the right, that's another survey. And so of APP ratios through our SCCM member registry. And so the ratio looks like the practice habit is around one to five to one to ten, give or take. But again, this is not data driven. This is just what we're doing right now. And so this has implications about what kind of model you set up in terms of APP inclusion. And so I'll end with what we do at Emory. And I think this is actually pretty interesting. I'm not sure how unique this is. I actually want to ask. But, you know, what we do at Emory is, you know, we have our ICUs and they're staffed by the various divisions, the academic divisions. So the MICU or the CCU might be staffed by folks from Palm Critical Care. The SICU might be staffed by folks from the Department of Surgery. But the APPs are hired not by the School of Medicine, but by a service line, a sort of a cross ICU, cross departmental service line in a centralized manner. And so that staffing, credentialing, evaluation is all done through a centralized body. And they do embrace the idea of standard work. So all the APPs do all the same things. So I can have an APP in my unit in the MICU at one of my hospitals. And then, oh, by the way, they're running low on staffing in the SICU across town. And the APP can just hop over there and do the same thing. And so I'll end with an example of my ICU. We have 18 beds in this hospital, Emory Johns Creek. Our staffing is a collaboration between academic divisions, including Palm Critical Care, the Department of Emergency Medicine. But our APP staffing is determined centrally through the Emory Critical Care Center using coverage ratios, the ratios we talked about. We embrace a ratio of one to six at Emory. They staff based on growth projections of that ICU versus other ICUs, and then the site needs. Our workflow, as I mentioned, is driven by standard work of the APP in terms of note-writing procedures. They all do the same thing at every site. And they're semi-autonomous. We embrace a high-revenue approach at Emory where the attending and the APP both see the patients. We collaborate on the care. Again, that's a little nuanced, and that's cultural based on our location. The old hospital I worked at before Emory 10 years ago, we did the low-cost approach where we all had our own patients. And as I mentioned, there's overlap duties. And so I'll end with that, but I did want to ask before I turn it back over to Dr. Bender for some Q&A, what of those that have APPs at their locations, do you do the high-revenue approach? Do you each see? Yeah. And how about division of labor? Yeah. So it's variable out there, and I think that's an important thing when you think about expansion and the role of the APP. Thank you.

advanced practice providers

APPs

ICU

healthcare teams

patient outcomes

education and training

APP integration models

quality improvement

teamwork