Well, I see many friends and respected colleagues in the audience today and I think that I just want to take a moment to thank you for your contributions to telemedicine and the background that really made this success possible. You guys did it. I need to directly compliment Jeremy and Ben for doing what was hard rather than what was easy. I can just tell you from my experiences every April 15th on just giving money to the government that doing this kind of a project is really amazing. And so to stand this up in the time they did, I think is really fantastic. I also wanted to thank Corrine for inviting me to give the easiest con presentation that I've ever been invited to do. And I would simply point out that it's widely accepted that during COVID there was increased critical care need and that the individuals, incredibly talented and dedicated individuals that volunteered in essence were very good, but they weren't perfect. And the systems that they worked in are really far from perfect. And therefore it's self-evident that the network could have done better. But I will go on and give you a little bit more drill down about this, a little perspective about things that I think going forward that we learned from this that informed us about what we should be doing and a little bit how we should do it so we can do it better, quicker and faster. When we measure effectiveness, you can take a look as Jeremy had done, some of his data will be in here. The total critical care need, if you subtract out the critical care needs that are not addressed by any TCCN and you subtract out the nets and inefficiencies or harms, that's an index of overall effectiveness. In the analyses that I've done, the quantitative analysis that I'll present to you, I'm going to ignore any inefficiencies. I'm going to assume, even though I'll show you some data that it might not be the case, that the NETCCN is equally effective to other forms of critical care such as the continuous model. I'm indebted to Dr. Geiling who authored the planning and organization chapters in our emergency, in our intensive care textbook. I think it's the single best resource if you're a non-professional disaster responder to get up to speed. He loaned this slide to me. In addition to the patient-facing materials put out by the CDC and a variety of review papers, this being a particularly well-written and recent one, there's general agreement that when you go to respond to a disaster that it requires, you can bring resources in terms of space, stuff, and staff. Now the NETCCN intervention really focuses mostly on staff. It does address indirectly space, but not so much stuff. Because it's really a staff-focused response, it's not as robust as necessarily is required to really optimally deal with disasters. We talk about telemedicine support. There are several things that we can do that can really be helpful and bring value. Those of us that work behind the camera have done all these things. One of the things is we can help recognize an evolving emergency. The earlier we recognize that the patient is starting to become physiologically unstable, the easier it is to deal with the situation. The longer you wait, the easier it is to recognize it, but the harder it is to deal with. So recognition is important. And I'll give you some examples where the continuous monitoring approach may be a little bit better than the on-demand approach in that regard. One of the things we can do is arrange evacuation. This isn't always possible, but one of the key concepts here is are we going to take Muhammad to the mountain, which is arranging evacuation, or are we going to take the mountain to Muhammad, which is things like transitioning to palliative care and a critical care intervention at the current site of care. If you're going to do the take the mountain to Muhammad, and a little bit even if you're going to just arrange evacuation, you still have to create and resource teams on the fly. And so one of the key things that the Netson folks and the continuous monitoring folks were doing is they're taking people who maybe don't practice critical care all the time or aren't comfortable doing so and creating conditions where they can be effective and comfortable serving their patients. So one of the things about activation that matters are several lessons that came from the Hurricane Katrina disaster. So this photograph of this young boy here was taken four hours after the storm had cleared. He's sitting in his house. He has his only remaining position, his teddy bear in front of him, and he's emotionally overwrought. And so many of our frontline providers, when they're faced with the possibility of an unexpected death or more patients than they can take care of, they feel not all that differently than this little boy feels. Now about this time, there were several interviews of people that had survived Hurricane Katrina. One of them that I remember was on ABC News, and it was a woman that was, her house was destroyed, but it wasn't completely unlivable. And it was several days after the disaster, and they were asking what things that the management associations were doing that really helped her and what didn't. And she said, you know, sometimes when people come around, they ask me what do I want to eat for dinner tonight, I just don't, you know, I just, it's just a burden to try to figure that out. Why don't they just bring me a pizza, because they can see that I'm hungry. So as we look at this little boy, you can tell that the little boy is hungry. You can tell that he's thirsty, and that he's going to get cold. But does he have what it takes to reach out and to ask the photographer to have, to share some of their water with him? So one of the problems that we found early on is that we had done some qualitative research about recognizing instability in our ICUs, and we found some surprising stuff. I think this has evolved over time, this is older, but the fundamentals are still true, and it's not all that divorced from observations around Hurricane Katrina. So in our ICUs, we had about 6.8 true positive alarms a day, the bedside staff were on top of five of them, but that was 1.8 times a day when there was an emergency that was evolving that the staff hadn't responded to. And about every other day, that happened when the intensive care team was actually rounding on the unit. And it's not an accident, because when the ICU team's on the unit, there's sort of an intense focus on one patient at a time, or the sickest patient, and it creates a situation where you like have a 400 square foot rug in an 800 square foot room. You're moving that rug around, but it can't be everywhere all at the same time. And some of those patients are left uncovered because the resources have been pulled away from them. The qualitative analysis that we did is we just counted up, because we knew how many phone calls people made, how many times they pushed the button to activate the system, and the continuously monitoring paradigm, and we recorded roughly 25,000 interventions. Of those, 24,426 were initiated by the offsite team, and 483 were the onsite team asking for help. So it isn't that they necessarily didn't know they needed help, it's just that in their workflow, taking the time to call for help wasn't as important as doing their other tasks. Translated what this means is that for every time that the on-demand system can recognize an opportunity to make a difference, there's 49 times when it misses them, because the onsite team simply can't activate the system. And I want to drill down. I normally don't tell anecdotes. I spent a long time in my career not doing this, but I decided today I really needed to do this. This is a case that was, Theresa might remember this case, it was a case that we served when she was, we were lucky enough to have her helping us out. This is a case of a 58-year-old, 400-pound man, and an alert came across our system that he was hypoxemic and tachycardic, and we turned on the camera, and we realized that the bedside intensivists and respiratory therapists were struggling to try to get him intubated. We had evolving respiratory distress, low oxygen saturations, this was in the setting of status asthmaticus. So one of the questions that I bring out, that I bring to you guys, is how can an on-demand service engage when there's no one available to make the request? Now as it turns out, as it turns out, the intensivist was somebody who we had trained before, so I knew him well, and I knew that he was a skilled intubator and a world-class bronchoscopist. So before I said anything, anything unhelpful like, how can I help you, or even worse, you know, do you want me to get somebody up here that knows how to intubate or something along those lines? The first thing I did is I decided not to speak. I called down to our OR staff and arranged for them to deliver a video laryngoscope to the ICU, including setting it up. And then I clicked on my microphone, I said, good evening, this is Dr. Lilly, your telemedicine support physician. In case you would find a video laryngoscope to be helpful, I've arranged for our OR staff to deliver one and to set it up for you. Would you like me to see if another experienced intubator is available to support your efforts? So here, this is the equivalent of giving the thirsty kid the water bottle and then asking him, is there anything else that you need? Because you've communicated that you understand their situation and that you're there to help. It turned out, possibly because I knew that I had a relationship with the intensivist, that he was more than willing to have someone come up and help him, because by the time I turned the camera on, he was already, you know, sweating. So I speed dialed the OR desk, I found out that the in-house anesthesiologist was going to be in the OR for at least 60 minutes. But I asked them, do you have a backup? And they said, yes, we do. Our chief anesthesiologist, who's our best intubator, is available, why don't you give them a call, give her a call. So I called her at her home, and she is so good at intubating that she almost never uses a bronchoscope. But she was the best intubator of the group, and she said she could be there in 20 minutes. So I asked the OR staff to bring a bronchoscope up to the ICU before the anesthesia chief arrived. It took 52 minutes with the anesthesia chief manipulating the video laryngoscope, and the intensivist manipulating the bronchoscope until that airway was secured and he was ventilated. And I can tell you from looking at their scrubs that they were sweating, and there was a big sigh of relief when everyone was convinced that the airway was secure. So let's take a look at the two models. So in the on-demand model, you have to learn from the bedside what's going on with the patient. In the off-site model, you have an electronic summary. You know his weight, you know his major diagnosis, and you know the history of his. You can look at the flow sheet and see what's been going on with his vital signs. In the on-demand model, the bedside provider has to tell you what they need. On the off-site model, because you have these other ways to know, you have ways to understand what the situation is, and they don't have to communicate that. In the on-demand model, a telemedicine provider calls the hospital operator to get the OR number. They have a delay, and when they finally do get them on the phone, they find that the anesthesiologist is not available and don't necessarily have the relationship or the understanding to say, oh, there's also a backup person available that can come in. In addition to being able to reduce the time to airway security for this patient, knowing the other patients in the ICU, the continuous monitoring team can say, oh, I've pulled the only RT out of this 10-bed ICU. There's a number of other patients in there who are unstable and need RT services. I'm going to get one of the RTs that's on the floor to come in and cover during the emergency. So I just think that when you look at the on-demand versus continuous monitoring model, that there's a lot of reasons that the continuous model is more effective at dealing with acute emergencies. So Jeremy had nicely shown that there were 4,560 hours of incremental critical care that this team provided, and that is amazing, absolutely amazing, and it's a great accomplishment. I think really he lived up to the commitments he made to council here at SCCM, and my hat's off to him for doing that. But I did publish that it was a rather small fraction of the total work that was done relative to meeting the emergency. So I think that this is something that Netzen definitely deserves a lot of credit for, but there are other people in the telemedicine community who stepped up and also contributed. So I want to make an analogy here about what we really learned from the COVID disaster. So what did we learn from the RMS Titanic? So this is an artist rendition of the Titanic going down, and the main message was that prevention was not enough. You had to be prepared. That is, that focus on hull design and iceberg detection were not enough. You also had to have enough icebergs. So on the bottom left-hand corner, you can see the individuals that designed the Titanic, and you can see a rendering on the right-hand side, you can see the office where the engineers actually did the design. Below the left shoulder of the main engineer there, you can see the picture of the RMS Titanic hull design that was reviewed when they went back and said, well, how come it sunk and what could we have done different? On the right shoulder is a picture of the lifeboats. So I wanted to just share with you that when they went over the hull design, they discovered there was a couple things that they might have taken care of and done a little bit differently. But the lifeboats, they didn't need to redesign them. There wasn't a single lifeboat that failed. The lifeboats actually were kept in the harbor, and White Star Line didn't get rid of the lifeboats. They deployed them to the other sister ships of the RMS Titanic, because if they didn't have enough lifeboats on them, nobody would go on those ships. So the equivalents here in COVID are the hull design is the equivalent of a focus on vaccines that are going to wipe out the pandemic, and the boats here represent doctors, nurses, and frontline critical care providers. So if we ask, what did the pandemic teach us about critical care? Did we learn that we need better support for intensivists, pharmacists, nurse practitioners, physician assistants, respiratory therapists, nutrition service professionals, and nurses? Or did we learn that the only thing we need to invest a lot of money in is better public health measures? And so I ask you, in these pictures at the bottom, you can see Dr. Fauci's effective here. You can even get U.S. presidents to wear a mask if he works at it hard enough. If you look at the bottom of this picture, this is public health. Where is critical care and health? How come Tim Buckman, or better yet, Jeff Drazen, who was the editor-in-chief of the New England Journal when this all went down and once told me that he was one of the five physicians in the country that could get the president on the phone, how come they're not at these conferences? And if they had been, would they have been saying things like, this ventilator reserve package that we have is more than enough to deal with the crisis, because all we need is stuff? Would they have said that we just need more money for ASPR or for BARDA, which is what the government funded? Because I'll tell you, not a single one of my critical care nurses think that more ASPR BARDA funding is what is required to better deal with these disasters. So what did we learn from the NET-CCN experience? One important thing was that setting these things up is really hard, and even when everyone wants to do it, it takes longer than it should. So one of the technologies that we now have is this eConnect technology. I was in a foreign country with a team that went in, and they were talking to the C-suite, and I couldn't have the language skills to understand. And while they were doing that, they installed this in the ICU, and when that conference was done, they had taken an ICU workstation, all stuff that they already owned, and they had set up a monitoring center in that amount of time. And that was a little less than what Ben had presented. I think that a policy that required all hospitals to have 10 percent of their licensed beds monitored with a network connected biomedical monitor that included a camera, a microphone, and a speaker would expedite the emergent deployment of telemedicine support where and where it needs to go. This is something that could be done at almost no cost to the government, incrementally over time, and can be done at very low cost, and why are we not doing that? AI-driven monitoring solutions are important. They're important because they reduce the number of false positive alerts. Recently published an article that's going to come out in Chess in just a couple of weeks that demonstrates that with 78 intensivists, 78 nurses, and 78 pharmacists, you could cover the entire country 24-7. Policies from the monitor can provide actual items and real-time updates to the emergency response centers, and I think that rather than sitting around a table and saying, well, what's going on in the neuro-ICU today, you'd have real data about acuity and involving patient need. Our telemedicine pharmacists, they need to be managing national drug supply stores. We got a billion dollars for, like, stuff we need, and how much money do we have to deal with drug shortages? Maybe we're not using our national resources the way we should. And I think that the sort of folks that support our critical care teams every day have the skill set that's required to both establish and to manage those resources, and at least in my hospital, they'd be used every day. And the worst time to run out of a standardized medication is when you have junior and inexperienced people managing your ICU. And so why are we doing that when we don't need to? We need to do critical care better. The Federal Disaster Management Agencies have focused on an American Red Cross restorative disaster relief model, and they sunsetted their support for the NEDSR. And this is literally the equivalent of taking the lifeboats off the RMS Titanic and burning them rather than redeploying them where they need to be. I'm convinced that it's not going to be possible either for the Department of Defense or for the ASPR organization to efficiently manage the mobilization of these kinds of resources. I think the center that needs to do this is the National Institutes of Health, primarily because they've successfully coordinated between the key governmental agencies, including the Department of Defense and CMS in the past. I just wanted to conclude by, again, thanking each of you for your contributions to telemedicine and to tell you, I think, that what you do every day really makes a difference. Thank you.

telemedicine

crisis management

continuous monitoring

AI-driven technology

critical care responses