Morning, everyone. Welcome to Hawaii. My name is Sue Stempeck. Thank you so much for- Try that again, Sue. That was a little weak sauce. Good morning. Woo! Aloha. So my name is Sue Stempeck. I'm a physician assistant. I work in pulmonary and critical care medicine at Leahy Clinic in Metro Boston. Thank you so much for joining us for this early morning session where we are going to argue with each other about what APPs should do aside from their clinical practice in the intensive care unit. We have experts from the perspective of their clinical practice managing these things. But as you all probably already know, there's not an enormous amount of evidence around the topics that we're going to present to you. So we will go through them, and we will have plenty of time for questions at the end. And we're happy to stay and talk more in the hallway, as usually happens with these types of conversations. So without further ado, I will introduce my colleague, Allison Wines. You've probably seen in the program, we're going to talk about education, billing, and the role of critical care triage. Allison. Good morning, everybody. I'm Allison, and I have nothing to disclose, except for I do teach multidisciplinary teams. So you can imagine where this is going. And we're going to talk about how APPs can educate you for the next five to seven minutes, including your residents, your fellows, your RN, and your support staff. So there are a variety of ways that APPs can educate. We're going to break down ways that we can perform formal education, bedside, informal teaching, procedures, and a lot of that institutional and floor knowledge. But one fun fact before we get started, learning is for everybody, right? We work now in multidisciplinary teams. Critical care is a team sport. Even clinics are teams at this point in time. So we're all kind of here, really, I think, if we're being honest, we're worried about if APPs can teach residents, right? That's where the conflict is. That's where historically we worry about if they can supervise, if they can teach. What does ACGME say? And yes, we will address that. But there's also RTs that need to learn. There's pharmacists that need to learn. There's residents. There's RNs. Sometimes in my role, I even teach attendings who rotate in and out of our ICU like once or twice every year. So learning is for everyone. And the more we learn together, the better our patients will do because we all need to be sharing that same knowledge and having the same goals for our patients. Maybe, there we go. So a lot of people say, oh, well, APPs aren't really trained to educate. Well, not every physician is a great educator either, but we'll focus on APPs. So yes, it's true, not all of us are great teachers, but some of us really are. A lot of us really are. Some of us have formal training either from a degree they may have gotten somewhere else in life or focused on medical education. Some of us have been trained in simulation like myself. Some of us have been trained to teach procedures. This makes us a really good choice for teaching lectures, for maybe doing M&Ms for your division, simulation training, procedural training. I know at my institution, I actually have more simulation and procedural training than any other staff physician in my division, except for one. I'll admit that. So frequently, I'm the one who's leading those or helping with those. So let's focus first on bedside or next on bedside and informal training. This is probably where I think you get a lot of bang for your buck and a lot of teaching and training isn't necessarily recognized. Let's start with patient assessment. I'm currently working with a surgical team and I will admit my attendings don't always actually touch the patient. So I'm the one who is in front of the intern. I'm the one who's in front of the med student actually doing a head to toe assessment, actually getting out a stethoscope. I think we forget those exist sometimes. I'm the only one on my team who knows how to use ultrasound. So I'm teaching those people informally at the bedside as we are seeing patients. I'm usually a person who can quickly decide what medication we're gonna choose just because I've been teaching or I've been treating these same patients over and over again. So instead of the intern having to guess or us all debating about it, I'll say, hey, I know this staff physician likes this drug. Let's start with that. Also, I think goals of care conversations are a really good place where APPs can teach other members of the interdisciplinary team. These conversations are hard. They take a lot of practice and who better to model after than someone who has those conversations weekly or daily or constantly and is really honed those skills. Modeling your behavior after someone who's an expert in that area will get you there a lot quicker than just trying to wing it yourself or by maybe someone, again, who is just rotating through and hasn't had one of those conversations for a long time. We do procedures as APPs and a lot of times we are the ones who do them the most. So again, a lot of times your APPs will be the one who's always in that clinic or always in that ICU. So their paracentesis numbers, art line numbers, central line numbers are usually a lot higher than maybe some of your residents, your fellows, and even again, some of your staff who rotate in and out frequently. So who do you want teaching that intern how to do a paracentesis? A staff who's maybe hasn't done one or two in the last two years or someone who does them several times a month. So think about that. I put CPR up here. I know that's not really a procedure, but A, you can bill it as a procedure and that is another skill that takes quite a bit of time to hone, to kind of control that room, to be able to take command of not only knowing what comes next in that algorithm, but also how to delegate who's doing what and really keep control as what you need to do. So institutional knowledge, I think, is another thing that APPs are really good at knowing what policies are on the floor. Can the nurse pull this line? Can we give this med peripherally? What culture is on that floor? Who do you call? Who do you page? Things that just really aren't written anywhere, but are just kind of ingrained in that culture. Here, I've really helped this resident out. Dr. Morris likes to know everyone's astrological sign when we're rounding, and I told this resident so he knew, so he didn't get yelled at day after day after day for not knowing if they're a Sagittarius because that helps us determine what presser to use, I think. I'm not really sure. So do residents actually like this? Well, they do. There's not a ton of evidence out there, but there is some evidence saying that residents who have been involved or around APPs do think that they do add to their teaching, especially when it comes to protocols and guidelines. Well, what about the ACGME? Everyone's so worried about the ACGME, so if you're worried, please see ACGME Frequently Asked Questions on their website, questions 35 and 36. 35, does ACGME allow other healthcare professionals to teach residents? Well, the answer is actually yes. If they're appropriately licensed but credentialed and they're non-physician, they can go ahead and teach these other professionals per the ACGME. Sorry, the slide is not working. Advance. There we go. So if you read the next answer to question 36, the supervision is where it gets a little more dicey. They need to be, residents need to be supervised by an attending, but if you read through this, some of it can be farmed out if you have these guidelines and blah, blah, blah, blah. So people still get concerned about the supervision portion of it, but guess what? Teaching isn't necessarily supervision, right? You can teach someone how to do a skill. You can teach someone about a guideline. That doesn't mean I'm in charge of them. I'm just sharing some of my knowledge with them. So, as you can see, I think APP should be teaching and don't forget to evaluate this session in the app. Thank you, Allison. Hold your applause. What's your sign? So this is, being assigned the con on this topic is a bit of a challenge, right? Because I think, I'm not sure how popular an opinion this is. So I'm gonna take a slightly different tack than saying APP shouldn't teach people. That seems a little bit draconian, but what I wanna focus on is the should, should and serve. So the title is, APP should not serve to educate the most multidisciplinary team. This is me. I'm at the University of Washington and my only disclosure is I work on a lot of teams, with APPs and they're all fabulous. But in the next 10 minutes, I wanna go over three reasons why APP should not be expected to routinely teach house staff in the ICU. There are some considerations in this issue I think we should pay attention to. There are some clear advantages to the APP role in teaching and we heard some of these just a moment ago. There's consistency in a lot of these teams, particularly if you're in a big academic center. Lots of house staff turnover on a given team. Many of the APPs that I work with are very experienced. And the ACGME not only allows, but actually requires interdisciplinary training in their common program requirements. Not necessarily teaching, but interdisciplinary care. And there are generally, as we heard, positive reviews of educational impact on the role or the addition of APPs to the ICU team, or I should say medicine team. These studies have been done in a variety of settings. But the question is, has that been done in evaluating the role of APPs as teachers specifically? And the answer is no. So the assessments that have been done, looking at the effects of APP integration into the ICU as well as other teams, this data actually comes from a general medicine ward team. They've not focused on the role of APPs specifically as teachers. If you drill down in the data, what you get is questions like this. So this was a survey of residents, total of 137, a year before APPs were integrated into a ward team and then a year after. And what you can see is that the question they actually asked was, is the workload appropriate for me as a resident to pursue educational activities? And before APP integration, only about half said yes. But after addition of APPs to the team, far more did. But they weren't asking, how were APPs as teachers? Did they free you up with the workload of the team to get additional education? And the answer there was yes. So there is of course a role for APPs to teach in certain circumstances, but we don't have good data necessarily to say that broad strokes, APPs are effective educators in all settings. And in particular, we're focused here on the ICU. So what does make a good medical educator? If there have been a number of studies, there's a great review here. This first one listed goes all the way back to 1909, looking at papers, evaluating, asking learners, asking teachers, what makes a good teacher? These are the top three in the qualitative analysis that they come up with in all of these papers. Number one, knowledge and skills. The other two, I would argue, could be a universal attribute of anybody on the ICU team, especially enthusiasm, big fan of that. But what about this knowledge and skills? And while many of the APPs that I work with are very skilled, all of my colleagues up here, incredibly skilled, very experienced, you can't take for granted that somebody joining your team, particularly as a newly minted APP, has the educational background that we all hope folks will have in time. And the data here comes from evaluation of overview of programs, broad strokes, nurse practitioner and PA programs, and then a generic MD who's gone through two years of clinical training as a med student, followed by a three-year resident. And this refers to the number of clinical hours spent in training that's required for the programs to graduate. Now, this doesn't include, of course, practice that, or clinical experience that this nurse practitioner had as an RN before doing their program. This is an overestimate, I think, of the difference. But the key thing is when someone comes out of an APP program, there is a wide variety of what their experience has been and how much time they've had, particularly in the ICU. Post-graduate programs are not widely available. There are a few out there, if you're interested for yourself or for team members. These folks here have fantastic resources about that. The Vanderbilt Bootcamp, for example, is one. But most critical care APPs actually don't receive advanced training during their required training before they accept their critical care job. It's a lot of it is OGT. And raise your hand if you are an APP and that rings true for you, or if you're uncomfortable doing that. You don't have to, but I see a few hands. Yes, thank you, thank you, okay, great. So you can't assume that folks are joining your program knowing the stuff that you hope that they're gonna be teaching your learners. And this is a universal statement, really. I'm pulling out here procedural experience because we talked about procedural experience as a particular advantage just a moment ago. You could draw the same graph, I think, for MDs just as well as APPs. This study happened to be one looking at mastery learning of APPs in a single center trial looking at, I think there were roughly 20 APPs in the study. And these were all experienced folks anywhere from two years to 10 years experience in the ICU routinely doing these procedures. In the initial evaluation, as you can see, not everybody met the mastery level learning criteria that then they demonstrated over the course of their training session that they were able to reach by the end. Now, like I said, I think you could draw this same graph for MDs. This is not to pull out APPs. The point that I'm trying to make is acknowledging that the fact that you are working in an ICU does not necessarily mean you have the expert skills to impart to your trainees. Now, the final point that I wanna make is that burnout is a universal phenomenon, right? A lot of the data focuses on the healthcare providers writ large. There's a lot of data out there about MDs. There is also data out there focused on APPs, not as much. When you look broadly at the literature, you'll see that there are three main buckets that lead the list of causes of burnout, right? Working hours, difficult relationships with patients that skyrocketed during the pandemic and their families, patients and their families, and then relationships that you have with coworkers and the administration of the hospital where things like the EMR. So assuming that APPs on your team are going to take on an additional role as educator of trainees runs the risk of exacerbating this working hours, the number one reason that folks cite for burnout currently. And so I think the assumption, again, this is the assumption that APPs are going to be joining our team as educators and performing that role. So we're keeping tight to time. So I'm gonna stop here at just a quick review. The three reasons I think we should not be expecting APPs to routinely teach house staff in the ICU. We don't have this concept very well studied in the literature in terms of efficacy. Although we all know fantastic teachers for APPs, it can't be assumed because of the wide variety of educational background that folks are coming in with and the paucity of postgraduate training programs for ICU or critical care. And then finally, remembering that the expectation that this is gonna be a part of your job, but generally not reimbursed, right? We all know this, may contribute to burnout. So be cautious about assuming this is gonna be part of your APP's job on your ICU team. And I'll stop there. Thank you so much, Dr. Morris. And we will welcome Dr. Rice and Megan up for the next session. Good morning. We know that everyone wants to talk about billing at seven o'clock in the morning in Hawaii. I know that's what brought everybody over here because that's clearly the most exciting thing that we do. So my name's Megan Kirkland. I am a PA by practice. I also am the executive director of advanced practice at our hospital system in Piedmont and I work mostly out of Atlanta, Georgia. I have no relevant disclosures. And we are going to spend a little bit of time kind of overviewing billing very briefly, knowing that this is a much larger topic than seven minutes and hopefully convince you that APP's should be able to bill. So there are a few different types of billing just as a quick review. So you have direct payment. You can register the APP with the payers under their own NPI number. They bill for their services. Those services are then paid out to, paid out under the APP reimbursement. There's incident two billing. This is predominantly used in the ambulatory setting. There are a lot of rules that are associated with this that again, I won't get into all of those in this scope of this discussion. But it allows the APP to see patients with an established plan of care if they're not making any changes to that plan of care and billed under the physician at 100%. There are split shared visits. This is predominantly used in the inpatient setting. When there is a medically necessary encounter that both the APP and the physician are performing a substantive portion of the care, then that care can be billed at 100% under the physician as well. There are documentation rules associated with this. Again, that we could get into in a more focused billing discussion. And then critical care billing, which is a time-based billing service. Can be additive within APP lines or physician lines and define charges under each of those specialties. So APPs can bill anywhere. So they can bill in the clinic, they can bill in our ambulatory setting or inpatient settings as well as in the ICU. And hopefully we will convince you that there are lots of benefits to APP billing as well. So we'll go over maybe just quickly four of them even though there I think are significantly more benefits to why APPs should be able to bill. In the ambulatory setting, APPs can see established patients like we mentioned that don't have a change in their plan of care and bill that at 100% under the physician as incident two billing. That then allows those routine patients to be seen by an APP which can open up your physician schedule templates. So and then if you're not gonna bill incident two, you can also still do direct pay in the clinic. And that you can kind of put your routine follow-ups with the APP and then your new complex patients that opens up that schedule template for the physicians to be able to spend more of their time with that patient population. We also know that there are a lot of varying schedules. So whether you have physicians that are rotating on and off service and spending their off service time in the clinic, whether there's vacation times that are very popular, by having a consistent APP in the clinic, you're able to still generate revenue while maintaining that flexibility of schedule for the rest of the clinic team. And then we also can have new patients set up with APPs. You can set criteria differently in your clinic depending on what types of patients you see. But that can decrease our wait times and improves access to care, hopefully decreases readmissions and makes patients a lot happier when they can be seen in six days instead of six months. APP billing also expands the reach of our healthcare team. So when you can allow APPs in the ambulatory setting, they can set up specialty clinic opportunities. So for our clinics, we have an interventional pulmonary APP clinic. They see patients pre and post-op. That frees the interventionalist up to be doing procedural things that she can focus on that only she can do. And then you are still generating your revenue from the clinic. Those patients are being seen in a timely manner and you're still generating your revenue from the procedural time of the interventionalist. In the inpatient setting, APP billing allows for revenue generation while you still maintain continuity of care. So again, with physicians rotating on and off service, you have a continuous presence of that APP that adds some historical knowledge, process knowledge within the hospital and still allows that revenue generation when the physicians are rotating off service. It can increase the number of patients that a physician can see. So you're expanding your reach of patients that you're able to see on a daily basis in the hospital. And it also allows for some physician flexibility for other tasks. So the APP can still man the floor list while the physician can open an afternoon clinic, can go to Bronx in the afternoon, can really give you a lot more reach to your patients in the hospital as well. APP billing can also increase department revenue. So in a critical care setting, we know that those are time-based billing codes. You can increase your opportunity for that 99292, that second level of billing, especially because your nocturnal presence in a lot of hospitals are manned by your APPs, that with APP billing, that time is combined. So for those patients that are critically ill or decompensating, you have a higher chance of reaching that higher level of billing by having your APPs bill in that combined time. They also have the capacity to bill for other tasks, procedures, consults, CPR, and then other specialty services that you might have that you can run 24 hours a day with that APP presence. And then they're able to bill, and they're a lower cost staffing model. So this is, again, probably a lot faster than like 120, a lot more detailed than 120 seconds, but APPs make less than doctors do. And the RVU reimbursement to physicians is often greater than what the RVU is, the work RVU is actually reimbursed to the hospital. And so you end up with the more productive a physician is, the greater your budget deficit can become when they're paid on a productivity model. They also can offer the same billable services at a lower cost. So quick math example, again, because everybody loves math so early in the morning. But if we take a 99291, which is your entry level critical care time billing charge, it's worth 4.5 work RVUs. For this example, we're gonna make a couple of assumptions. We're going to assume that that 99291 took you one hour, because that's just a little bit easier for math. We're also going to assume that the physician is paid at $60 per work RVU. So for a physician to bill that 99291 in that one hour of time would cost $270. That would be the reimbursement to the physician for that hour. An APP, if you take an average salary of $150,000 and 2,080 hours of work per year is $72.11 per hour. So for that same charge of a 99291, you pay an APP $72 and you pay a physician $270. So it does end up in your salary and wages bucket being significantly more expensive. Dr. Rice will try to convince you that APP billing is often reimbursed at 85% and that is detrimental to a budget and that's a lot of the reason that we get why APPs should not bill. But even if they are reimbursed at 85%, though some payers do still reimburse at 100%, it's at 30% of the cost. So you're still in a win from a financial standpoint. Thank you, Megan. Dr. Rice, you're up next to try to prove Megan wrong. So what's worse than talking about billing at 7 in the morning is talking about not billing at 7 in the morning. And I got this talk mainly because, I don't know if you guys can see it or not, but it says sucker across my forehead. And that's why I got this talk, because nobody else really was voluntold that they could do this talk. So I'm going to walk through maybe why APPs should not submit bills for their services and I'm going to focus on just the ICU. If you have a post-ICU clinic, then some of the ambulatory stuff may be pertinent to you, but I'm just going to focus on their care in the actual hospital in the ICU. That's me. Here are the objectives. First we'll go through some billing models. Megan already talked about some of these, but we'll talk about them a little bit more. And then I'll try and explicitly go through pros and cons for models where APPs bill versus not bill. Conflicts of interest, I actually am a consultant for some companies, none of them do billing, none of them do APPs, so that's not a real conflict. The red one I think is a conflict, which is I'm a director of an ICU where our APPs actually do bill for their services. So you sort of see, you know, where I'm coming at from this standpoint. Megan talked a little bit about this. There's some common billing models. One is that the physician is the biller. All the bills go out under the physician's name and the physician is the biller. Another one is that the advanced practice provider bills for the services that they render and the physician bills for the services that they render. And it's almost like they're separate and they go out kind of separately. And then there's a mixed model, shared split billing, which my coding folks and billing folks have tried to explain to me maybe 40 times, and I'm still not sure I entirely understand it. But here's what I know with it. The physician and the APP have to be in the same group. That group isn't very well defined, so it could be, you know, a big group or a small group or employed by the same employer. They both have to have performed a substantive portion of the visit with the same patient on the same date of service. And the substantive portion involves some of at least history exam and medical decision making. So you can't do a shared model if the APP was the only person that saw the patient that both the MD and the APP have to have seen the patient. But if both see the patient, then you might be able to do this mixed shared model. And Megan, in preparing for this, told me, and I think she's right, although don't tell her I said that, that that also varies a little bit by what region and what state you're in. And so, you know, I'm in the state of Tennessee. So if I say something that sounds funny to you, check with your billing people because it might be that we're operating under slightly different rules than you are. So in the common billing models, when the physician bills, the physician bills at 100% reimbursement. When the APP bills for just the APP services, they bill at, at least in our model, 85% of reimbursement. In the mixed model, this can actually be reimbursed at the physician's billing model, as long as the physician has more time than the APP, with the rules that we already outlined on the previous slide. I'm an academic guy, and so I was like, hey, this will be easy, we'll just go to PubMed and find some articles about billing. There are five actual articles about billing. In case you're wondering, if you put COVID-19 in that top number, you get like 34,000 articles. So billing's a little behind COVID-19 in the academic publication world. If you go through these, none of them actually talk about billing models, and they talk about some procedure stuff and advanced practice providers, but none of them billing models. And they locked my slides on me this morning, so this slide was supposed to be the Google version of the search, because you guys will be like, well, nobody's doing like, you know, randomized trials on APP billing, so go to Google. If you go to Google, you get lots of hits, and it's all like people trying to sell you courses to improve your billing. None of them actually tell you anything about like, you know, the rules online or anything like that. So here's the pros for APP billing, and Megan talked about a lot of these. They can bill for individual work and capture work that might not otherwise be billed, especially if the physician isn't around for that. They can bill for their procedures. They have the ability to expand the billing beyond the patients that were seen by the MD, so they can extend the reach of the group and see more patients, even though the MD may not have the time or the ability to see those patients. And then, and I think this is important, and honestly, this is one of the big reasons in our practice why my APPs bill, is because it increases their autonomy. They are known to be billing providers, and they then have the ability to do lots of things in my ICU, because they can bill for it, and so therefore, the hospital says, sure, we'll let them do that, because they can bill for it. There's less of a feel that the APP salaries are an expense or a cost of doing business. There's a return on investment, and you can actually see some of that return on investment. Cons. We talked about the 85% rate, so they don't bill as high as MDs. They may decrease the efficiency of the overall practice. If you're seeing everybody together, and then you're doing individual billing, that may decrease the efficiency if you're going to do the mixed model and bill everything together as one. I don't know if this is a huge con, but they have to have their own MPI in order to bill. I get asked periodically what my MPI is, is how well I do billing. I don't know my number. I always have to go and look it up, because it's not a number I regularly use. They also have to be credentialed. They have to be credentialed at your institution. They also have to be credentialed with insurance companies and approved as a provider for those insurance companies. So there's some paperwork. Most of us, I suspect, are not the people doing that paperwork. And then, if in that mixed model the APP bills more time than the MD, then the APP takes over the MD billing. There are easy ways around that, at least we've found easy ways around that. The split-share billing model allows the group to bill APP work at 100% reimbursement rate, and it synergizes the work between APPs and MDs. But it requires both parties to have this substantial work in the evaluation and management. We already talked about that, and we already talked about the last one, which is rules can vary by region and state. So ultimately, I think it comes down to this. Sometimes you have to do what's best for you in your life, not what's best for everyone else. And I think this is what billing is. It's what's best for your practice. How does that fit into your practice? What's the best way to incorporate it into your practice? So I don't think you can make an argument one way or the other. I think the argument is figure out what's best for your practice and do what's best for your practice. Thank you. Thank you, Dr. Rice and Megan. Time for some voting again. So let's see a vote for APPs should bill for their services, specifically in the ICU. I know Megan talked about a few settings, okay? And how about should not? So Dr. Towne and I are going to flip the script a little bit. As you've seen to date, we've had our physicians arguing for the physician side of the argument and our APPs arguing for the APPs should. Dr. Towne and I are going to do the opposite. And so I'll invite Dr. Towne to begin, assuming I can see the slides. There we go. The title's wrong on his. He's the pro. I am the pro. Let's just get that straight. No. Hi. My name is James Towne. I'm one of the pulmonary critical care docs at Harborview Medical Center in Seattle, Washington, for the University of Washington. My title or my topic here is APPs should act as critical care triage officer. That's me 10 years ago, and I have no disclosures. I am the director of a medical ICU, and the fact that I'm in an ICU is where APPs regularly are valued and trusted colleagues, and they're here right now watching me. They didn't know I was doing this talk, though. Okay, so when we talk about triage, I think I like to step back a little bit and do a big picture question. One, I agree with Dr. Rice, is my favorite kind of talk when there's no data so I can say stuff. But triage, when we think about it, like back to the roots in agriculture medicine, there's a French verb. I don't speak French, but what I'm told is that it means to sort. And there's some people who think etymology also includes into three groups, and historically that was thinking about agricultural products, like wool, like sheep, like coffee, and you would try to separate the good beans from the bad beans. And then it got flexed into medicine in sort of Napoleonic times when they would run these massive crusades to parts of the world where they had no supplies and no resources, and they would try to decide which injured soldiers deserved medical treatment. And that depended on the goal of what that soldier was there to do, to win wars. It didn't matter about the soldier. And so it really depends on the goal of your triage system. Why are you, what are these resources important for? As we have all seen in the past couple years of COVID, there's a lot of modern healthcare implications for triages. One of the things, the implications is that you're saying that resources are finite. We can't all have everything. We have to decide who gets something and who doesn't get something. And so we are going to prioritize or decline treatment for some people, but not others. And hopefully these decisions are based on ethical and moral frameworks that are acceptable to the community. And that was a big problem with COVID is who's deciding for everybody else. So now when we scale that down into an intensive care unit, we want to say, well, you know, in the ICU, we often get asked to triage a lot of different kinds of calls. And so I work at a large county hospital that takes a huge referral from outside. And so I get triage calls from places in Alaska and places in Montana, and I get a call from people down the hall. And so you could be triaging from the emergency department, from your inpatient wards after rapid response. Maybe somebody is out of the operating room with a PACU and they need to know where they're going to go next because of this or that complication, even though they went in for this or that surgery. And then again, this idea of an external facility transfer. All of these roads lead to key questions for the triage officer. Will this patient benefit from critical care right now? What type of placement matches that patient's needs? What are the usual pathways we do? I mean, this is not the first day we've done this, right? And what's the current state of our system? I think that's an important consideration to think about. And what's the triage officer or the hospital's role in the hospital and region? And so if I'm running a private practice hospital in a very densely populated metropolitan area, I have different goals than running a large county hospital in a very rural area. Those people have entirely different hospital goals and cultures. And at the end of the day, then you make this decision. What am I going to do for this patient? So question number one, will this patient benefit from ICU admission? I think it's, when we talk to people who don't work in critical care, their answer is always, of course, why the F not, Dr. Towne? So take this patient. But in fact, it's not true that not all patients benefit from ICU admission. There's a recent provocative paper by an EC where they looked at patients with respiratory failure versus sepsis. And they found that patients who had sepsis that went to the intensive care unit actually probably didn't get that much more benefit from being in the ICU. I'm sort of condensing a whole lot of that paper into one sort of snippet for you guys right now. But patients with sepsis tended to stay longer, and they tended to have similar outcomes versus patients with respiratory failure tended to have way better outcomes going to the intensive care unit. And I think there's a lot of reasons for that, and it was many, many hospitals that were all lumped together. So it's not just one system, but it is a provocative way to think about things. We also know that there are patients who are too sick to benefit from ICU care. You may do everything you could possibly do, and it may be within that patient's goals of care. But in fact, that patient may not derive a benefit from there, or the patient's family may not. On the other hand, there are patients who are too well. You bring them to the ICU, and the exact same thing happens. Same monitor, same treatment, etc., you get the point. However, if we delay ICU care or we mistriage patients to the wrong initial location, those patients actually do worse. And so this is a really critical step. Early, accurate triage is associated with better outcomes than late or missed or delayed triage. And I think to wit, when patients are sent to an acute care ward and they deteriorate on the ward, they detract from the care of the surrounding patients. And so that's another missed opportunity to take care of those patients who weren't even involved in your triage calculus. And of course, context is important. So my situation is different than your situation, like we talked about up here. If I said I have five followers and I'm on Instagram, that's pretty shitty context. If I said I have five followers down a dark alley, that's scary. So- It all matters where you're talking about. An important thing in this day and age is bias in triage decisions. So as we talked about, it's not clear that benefit is there in all cases. And so lots of things, when we study this in literature, things matter to the triage decider. What's my ICU capacity strain looking like? How crazy are things? What's the age of the patient? What are the local factors? How do I interplay with other systems in my hospital? They all impact decision. What is less well-represented, I think, is the idea of unrecognized bias. And I think that's a huge factor in what we do in medicine these days, is whether I think that person will get better or not. Implies what I was already thinking about that person or about that situation. Someone who comes in HIV positive, not taking their ARTs, endocarditis. That person is gonna get a different look than the person who comes in, little old lady fell down on the ice. I think they're gonna be treated differently, unfortunately. Another thing that I think will bias a triage officer is, is their duty to the patient or to the system? Or is it the system and the patient? Who's driving their boat? So the system may value optimizing resource utilization. They want the most widgets for the fewest dollars. They want this thing to be a reliable, robust process that works 24-7, all the time, with consistency. And you wanna reduce the workload for the rest of your team. Whereas the patient says, I want the best care for my needs. And I actually want that triage officer behind the scenes to facilitate the process, to smooth things over, so I don't actually know anything happened. And it's maybe a layer of clinical support. That person's sort of aware of this patient, so should things go south, they can step in. And ideally, we're all achieving the best outcome by synergizing those things. So in broad strokes, I'm actually not gonna answer the question. I'm gonna do like what a savvy debater does and not answer the question you ask me. It's politics season. And I'm gonna say, triage is not a person, it's a process. And so what we should do in this context is we should develop clear guidelines and pathways with the relevant stakeholders. That means all the big leagues from all the services in your center get together in a room and you do some horse trading and you figure out what's the way we can make this system smooth and operate for everybody most of the time. Not 100% of the time, but definitely most of the time. And you need to measure those outcomes. You should develop a system that learns and is modifiable. And it should involve clinical judgment plus the EHR. And I think this is actually a very interesting place for AI to get involved because this is involving some element of data and some element of human cognition. And that to me seems like a good opportunity for machine learning and AI systems, which I didn't see any literature on, actually. I just coined the 70-30 rule. You guys are welcome to take it home with you. And that's most patients can be categorized easily. And when they take hypothetical critical care patients and they give them to intensivists to do surveys on and say, would you do this or that with a patient? There's only agreement about 70% of the time, 60 to 70% of the time, which means that A, there's very little critical care. And B, there can be different right answers. And not everyone's gonna agree on the same thing. But those borderline cases is where you need judgment, you need collaboration, you need to share a vision. And I'll put out the alternatives because we always talk about risk, benefits, and alternatives. So without this system, option one is what I would call the free market. And this is your emergency department calling neurosurgery, medical ICU, and neuro all at the same time for the ground level fall with a subacute subdural that's not gonna get operated on, all right? Who's gonna take this patient first and get him moving? That's free market. Number two is dictatorship, okay? I am going to tell you what to do. Okay, no choice, I'm just gonna tell you what to do. And then last, chaos, right? Old Testament, wrath of God type stuff, fire and brimstone, rivers boiling, the dead rising from the graves, dogs and cats living together. That's the alternative. That's who it's going to represent. So what I say is triage plus APP equals value. The hospital says, great, this is efficient and reliable. This is cost saving. And in fact, that person can also be an arm of the quality improvement system of the hospital by being nestled within that really critical process. The trainees say, this is awesome. As we saw earlier, this can offload some of the work. Instead of being called down to the emergency department in the middle of rounds, I can actually think really hard about my patient and learn better and focus on rounds. And the APPs can dive in and say, you know why this patient went to our service instead of that person? And I'm gonna teach you about the systems of care in this case. And the patient says, I get a free second opinion. I wasn't aware of it, but I got a free second opinion who maybe came in and say, you know what, in fact, this patient has in fact has sepsis. More fluids and antibiotics were needed, not treatment of this other disorder. Smoothing the process and adding a layer of support to what could otherwise be a difficult situation for them, so thank you. Thank you. All right, well, what I didn't tell Dr. Towne is that I do speak French. So while he defined triage quite well for you, now we're gonna talk about the who based on his lovely framing of the process. So this is me. My disclosure, I should have done what Dr. Rice did. My primary job in my critical care practice is doing triage, and I'm gonna try to argue to all of you why I shouldn't have a job. I don't know what just happened there. There's a poll, that's weird. I'm gonna skip it. Okay, I'm gonna do- That's what happens when you upload your slides to her. Right, planning ahead is bad. Okay, so I'm gonna ask a question by show of hands. For those of you that work in a critical care setting right now, who does triage in your organization? The attending physician, okay? A fellow, lots of fellows. A resident, not too many residents. APP, any mixture of this group? Okay, so we have a lot of mixture. That's great, actually. It'll spark an interesting conversation. Okay, so what is triage? I don't need to redefine this for you. I think Dr. Towne did a great discussion about that. Patients from anywhere who may need critical care. And then there is the question in your organization of, is this to specific ICUs? Do you have a separate process for medical ICU triage, for surgical ICU triage, for cardiology ICU triage? Do those talk to each other? Do we think about resources in one setting and lack of resources in another under your same roof? Does this sound familiar to anybody? Can APPs influence that independently well enough? I'm gonna say no. So some of you may have seen this study from, this was done in Vanderbilt with Dr. Rice's team. And Whitney Gannon led a study to demonstrate a pretty different set of outcomes in patients who needed ECMO when there was capacity in the hospital versus not. I framed this for you to say, this is a really important problem, and we are going to affect patients' mortality by the decisions that we make every single day. And I agree with Dr. Towne, it is a 24-hour problem. Is there any evidence? Well, I think the people who have written about this would say there is, but I think the organizational context completely ruins any evidence you can find. Because it's about resources and setting, and as we all learned in COVID, a bed doesn't necessarily equal a staffed bed, right? So there's this context of acuity, how many nurses you have, how many respiratory therapists you have, how many physicians you have. However, this work through SCCM does frame a little bit of guidelines for you. So for those of you who raised your hand that this is something that you participated in your organization, you should probably read this to see if it applies to you. And I think this framework makes an assumption, for example, that you have an intermediate care unit in your hospital. Does everybody? Certainly not. And so I think some of these pieces of work that bring together well-resourced systems sometimes forget about those of us who have hospitals within our system that don't have all the fancy things. So what does it take to do triage? Again, I will agree with Dr. Towne that this is a really, really important process and it requires, I'm gonna also quote Dr. Morris, it requires a depth of medical knowledge. So do our APPs universally have that depth of medical knowledge? I'm gonna say maybe. Is maybe good enough for that individual patient? Is maybe good enough for your hospital administrators? Probably not for something that's this important. It requires system thinking every minute of every day when you're in front of that patient, system thinking. When you're on the phone with Alaska, system thinking. When the helicopter is bringing you that unassigned patient, what's your bed in which unit and how can you move the resources around before the helicopter lands on the helipad? The other interesting challenge that we all learned a lot more in COVID is does this cross over some type of political line, a state line, territories, sometimes even countries depending on where you practice and are there telemedicine implications for providing clinical opinions, advice over the phone? Do you have a license to do that? This requires research, management skills, communication. I think the last point is the most important. You need some street cred to do this, right? You're hanging up the phone, you're calling the chief of some ICU in your building, you're calling a rural ED to say, I'm sorry, we can't help you, we can't take your patient right now. That requires some experience, some guts and some steel in your stomach sometimes. The expertise is too variable for APPs and how can you come out of fellowship never having done this? How can the APPs steal from fellows, the medical training required to do this? We can't impact that education. A new physician joining the practice in which I work clinically who's never done triage well is completely useless in our tertiary care hospital. And sick not sick is maybe the most important skill that we can learn as critical care clinicians. And if we remove that from our fellows, we have created useless attendings. So relationships with those referring providers and the customer service approach to this to make people comfortable is also really important. And you have to be able to multitask. So this like I must concentrate on teaching rounds and don't take my teaching time away, come on. Real life, you have to do it all at the same time. So I don't think that I'm qualified to do the clinical job that I do every day. So please, now it's time to vote. So we're gonna vote on this one and then we'll take questions for all of them. So thank you all for attending.

APPs

critical care triage officers

debate

effective triage

clear guidelines

collaboration

AI in triage decision-making

biases in triage decisions