Awesome. Thanks for having me, and not break the mic. All right, so Dr. Oktober already told you about me, and a lot of the work that you'll see, we were fortunate to receive funding from several nonprofit societies, organizations, no commercial entities, so we don't have any other disclosures or conflicts of interest to share for this work. And like with all education talks, we have objectives, so my goals over the next few minutes are to develop common digital technology terminology. A big issue in this space is a lot of people say a lot of different things that mean different stuff, so try to get everyone on the same page a little bit. We'll emphasize where digital training tools fit within critical care training and education, and then explore some applications of digital technology in the critical care training and education arena. So let's start with what we're talking about. So when we say virtual reality, this is when your experience is being totally immersed in a purely virtual world. If I turn the headset off, you're in complete blackness, running into walls, turn it back on, everything around you is virtual. Augmented reality, unfortunately the thing that everyone knows best is Pokemon Go, so that's when your real world is enhanced by digital content, but you're very much aware you're looking through a screen, and your only way to interact with that digital content is through the screen or the device. And then mixed reality is a system that combines the real and virtual objects seamlessly, so this is what everyone thinks and hopes and imagines for, as I'm in my space and there's a virtual patient that I can touch and interact with with my hands and devices, and it'll respond to me. So that's what we mean by mixed reality. So what I'm not here today is to say that digital solutions is the end all be all in the answer to all of your problems and educational gaps that you can go throw away all those mannequins and everything will be great. The key is finding the right tool for the job, and we've been limited with the tools that we have available and often are trying to fit a square peg into a round hole when thinking about a learning gap. So digital technologies offer us some solutions for nuance and potentially to be able to better target educational problems or gaps. But it's important to not just walk blindly into the new, just because it's new and shiny and cool doesn't mean it's the right solution. So hopefully everyone in this room, whether you practice with adults or kids or what your role is, it's somewhat uncomfortable watching this child working, some comfort knowing that Ken's hand is there listening with his stethoscope and keeping a close eye on him, but that this sense that we all have in this room that this child is sick and is potentially heading the wrong direction isn't innate. You're not just born knowing that. That comes through experience and is not an experience that all of our learners and the people that we practice with have. So how do we learn that? So Kolb's learning cycle is a great foundation to think through how we learn these clinical skills. So you start with this concrete experience that you're an early student in the room assessing this patient, and then you do something. No one's around, you're in charge, you say I'm gonna feed this kid and then see how that goes, and you say well that didn't go so well in your reflective observation, and then you conceptualize and try to make sense of what just happened. Why did that not go well? What was I missing? What should I have been thinking about in that situation? This is also where experts come in, so those faculty facilitators that come and talk to you about your patient, talk about your plan, point out those findings and next steps, and then you apply it the next time with the next patient. And so this is how all of us through training have learned to do what we do. When we're thinking about how to train people specifically for respiratory distress, there's a lot of different approaches that we can consider. Unfortunately, they all come up short. So computer-based mannequins are really fantastic for a lot of things, but when it comes to respiratory distress, they lack the subtlety and nuance to show a lot of the exam findings that we know are key for cuing into a patient who's decompensating. I can't pay kids to come in and pretend like they're in distress, so limited in the pediatric space in general where standardized patients are helpful, especially for more acute pathologies and for younger patients. So what are we left with? It's through ongoing patient care, learning through the care of our patients, how medicine has done it for decades and centuries. The problem is that we know over the last several decades bedside teaching is in dramatic decline, and that's for a lot of factors, and I'm not here to solve those factors today, but we all know and are aware that the time supervising experienced providers can spend at the bedside with novice providers is dramatically less, and so you're left with this blind leading the blind scenario at the bedside. Additionally and appropriately, hospitals over the last two decades have really ramped up their safety culture, and so lots of systems and processes are in place to rapidly identify decompensating patients, shunt them off to the ICU. Unfortunately, most of those go around our junior learners in a quest for efficiency, at no fault of the folks in charge of that, but an unfortunate byproduct of protocolization and efficiency and automatic triggers. So our solution was, can we replicate this in a safe environment that's realistic and that's with immersive virtual reality? So you have a learner and a headset that everything they see is virtual, and this is a recreation of one of our inpatient rooms, and into these environments we can put all of the normal hard equipment in the space, we can put patients with various findings, and we can put actual other team members to interact with. We can then control the exam findings, so you can go from no distress to signs of impending respiratory failure, so all of the subtle retractions are not so subtle to us, but subtle to our learners, to show and make those illness scripts in real time. When is a student ever going to be able to stand there at a bedside with a child that looks like this for 5, 10, 15 minutes trying to connect the dots? Someone's going to come in and whisk them away, and they're going to lose that opportunity to build that illness script. With immersive virtual reality, we have that opportunity to recreate bedside learning experiences. So what did we find? So as you're doing any of this work in this novel technology space, you have to start at the bottom of Kirkpatrick and work your way up. So what we looked at first was the student's perspective on whether they felt this was helpful, because if your learners don't feel like this is a valid or useful exercise, you kind of can full stop there in terms of their engagement and moving forward, and so we had them rate themselves on the RIME scale for the assessment and management of respiratory distress, and across the board, learners perceived significant improvement. Some questions are raised by those third-year med students that start with the assessment of manager, but the findings are still valid in the sense of that they felt that the curriculum had an impact, and they had significant impact on their ability to assess, reinforce key assessment components, and that their perceived impact on the assessment of future patients. That's all well and good, but what we wanted to get at is, were they actually taking away any learnings from this session and able to apply it forward? So could they show knowledge application of what they gained? And so for this study, we took third-year medical students on their pediatric clerkship, their last mandatory pediatric experience. At the beginning of the rotation, they were randomized to control or intervention. Control got a standard curriculum where, with high-fidelity mannequins, talked about respiratory distress and impending respiratory failure. The intervention group got an additional 20 minutes with me in this virtual environment, assessing these virtual patients. And then at the end of their clerkship, we had them watch videos like the one you saw in the beginning, and write in free responses what do they see, what do they notice, what do they want to do. And what we're looking for is, did they recognize in a patient with impending respiratory failure that they needed an escalation of care? So that was saying things like, call the PICU, needs positive pressure, consider intubation. And so those students who were in the control, who get no more mandatory pediatric experience, even if they're going into pediatrics, it's a flip of a coin whether they could recognize that that patient needed an escalation of care, 46%. After 20 minutes in virtual reality, the rate went up to 73%, correctly recognizing that that patient needed help. Now, they're third-year medical students, they don't need to know how, they don't need to explain the physiology, but from a benchmark of safety, released onto the wards, that if they saw that child, they connected the dots that this child needs help, is a huge safety benefit. So that's all well and good with medical students, but can we target more experienced providers, providers dealing with greater complexity? And so we had worked through our system on how to replicate respiratory distress, but if we could figure out how to embed signs of poor perfusion, then we can do shock. And regardless of the ideology, fortunately for us, shock looks like shock, whether it be in the CIC for low cardiac output, on the surgical floors, from hemorrhage, or in the PICU for sepsis. And so we were able to do that, so the difference between the last video and this one is five years of development expertise, so you can see the environment seems much more realistic, and the patient also has increased in their complexity. We can now put multiple users in the same virtual reality environment, where they can interact together taking care of a patient, performing tasks, discussing the findings. Importantly, this rolled out during COVID, and those two learners don't need to be in the same space, so they can remotely interact in this virtual environment from across campus, across town, across the country. And so this allows them to work through that phase that we traditionally haven't had much to do for in simulation, which is the lead-up to decompensation. We have lots available for once the patient dies to practice CPR and compressions and tasks, but we've had nothing previously that could show decompensation to allow us to check cap refill, to feel the change in pulses, to notice that the patient's mental status is changing and how they respond to questions, and to perform the tasks at the same time to accurately reflect the cognitive load on our learners. And so for this study, again, we had to start at the baseline because this was a new, this was a new experience, and we needed to know what we were working with. So we looked at participant recognition of sepsis and how long it took that to happen. We did that through the SAGIT technique, which I'm happy to explain to anyone later, but it's a way to assess situation awareness at various time points. And what we found, which hopefully isn't surprising, is that those with more experience recognize sepsis sooner. And you could say, well, that should be obvious. Why waste your time doing that? Again, we're creating new environments, new training experiences, and so if this was overly dramatized, then novices and experts would be like, oh, that's sepsis from moment one. If it's not realistic enough and its experts aren't figuring out its sepsis by the end, then the platform is useless to be able to show, train new skills, assess performance. So now I'm going to talk a little bit about what you can do with all those mannequins that you have in place already and how digital technology can help increase the impact. So we, through an amazingly talented team at our institution, can take real-life people, this isn't a patient, this is my mentor's son, take them and digitally recreate them to the proportions of our available mannequins. And through that, and the layering of various exam findings through a lot of fun platforms, create a new patient that seamlessly layers over the mannequin. So you put on your headset, you can look around the room, you see your colleagues, you see the equipment, you see all the other people you're working with, but when you look down at the mannequin, there's a digital representation of the patient is tethered to that space. And so then we can put people into team-based training and see how they perform. So you can see they're wearing these headsets while going about their tasks. So the bedside nurse is hooking up a rapid bolus, the respiratory therapist is assessing, and as the simulation goes, we can take the patient from looking like this, add their vital signs, start to change, have the scenario progress. So we can adjust their cap refill, their modeling, their cyanosis, we can change their mental state and have them look something like this when they're doing this scenario. And so what you find is that their performance dramatically changes, and their focus dramatically changes. So all the nurses who watched our learnings are just staring at the mannequin, waiting for the vitals to change. Instead, everyone's looking at the patient, everyone's talking about the patient, they're cued into those findings, and sometimes they're responding to the patient prior to the monitor. And so what did we find? So when we looked at this training scenario, we had these interprofessional teams do a mannequin-based simulation right before this AR-based experience, so that we had a standard comparison for what they were looking at. And not surprisingly, everyone found the AR sim better than reading or didactics, but what was more interesting is for this assessment of clinical decompensation, they felt that it was more impactful than high-fidelity mannequin and standardized patients. When we looked at their ability to recognize various findings, traditional training, so mannequin-based, there are a lot of folks who felt that they couldn't effectively assess mental status, respiratory status, and perfusion status, but with the AR enhancements, dramatically increase in their ability to assess those findings. Similarly, with determining when the patient needed escalation, similar findings that the AR enhancement allowed them to better assess when the patient was hypoxemic, when the patient was in shock, when they were apneic, when they required CPR, and globally that the patient was decompensating. So that sounds awesome. So why haven't we rolled that out at every institution and do all our mannequin-based training that way? Well, as like any new front that's kind of pushing the boundaries of what technology can do, we had a ton of side effects. So for the docs or the team lead, life was great. You stand in one place, you look at the room, the AR enhancement was fantastic. For our poor bedside nurses who were running around the room grabbing equipment, changing their view, lots of side effects with this early platform in terms of blurred vision, difficulty concentrating, disorientation, and dizziness. And so I present this not to discourage and to say that AR is not the solution and is not going to be helpful, but to show that what our responsibility is as educators and people pushing this work forward is to again not just sprint towards the new and dive in headfirst, but to take our due diligence and say, is the technology ready? Are the benefits outweighing the cons? If everyone leaves your simulation nauseous, then the likelihood of them walking away with new skills or competence isn't going to be very effective. But we're hopeful. There's new headsets that just came out that are sitting in my office that may be a lot better. So more to come there. So where do we go from here? There's lots of different directions we could take and there's a spectrum of problems that we can address through digital technology. So there's those competencies that happen frequently that a controlled environment for practice would be beneficial. So those are things like respiratory distress and shock recognition to those rarer events that have significantly significant ramifications if not navigated appropriately. So that's moving towards cardiac arrest in the pediatric environment. And it's also important is how do we pull in the rest of our team? So through the use of multi-user platforms, can we bring in all the layers of respiratory therapy, nursing, physician, and trainees at different level to have the learning be tailored to their needs? And so what I want to leave you with is this kind of gut check of who our learners are. So the year 2000 is the year that our current first-year medical students were born. In 2004 is when our current nursing students were born. And so they have lived their entire lives with digital content at home and in school. And so we are painfully behind in terms of how we think about learning and our new approaches and continuing to rely on stuff that we did in the 1800s as opposed to stuff that's been around for several decades. So with that, thank you for listening and happy to answer questions at the end.

digital technology

healthcare education

common terminology

virtual reality

augmented reality