Hello and welcome to this Thought Leaders Session focused on the future of critical care. We really have two outstanding expert discussants to facilitate our conversation today. Dr. Rebecca Aslickson and Dr. Michelle Gong will provide the infrastructure for our discussion around the future of critical care. Our panelists today, along with Heather Meeson, are co-chairs for the Society of Critical Care Medicine Future of Critical Care Task Force. Dr. Aslickson is an Associate Professor in the Departments of Medicine and Anesthesiology at Stanford, and she is board-certified in critical care, anesthesia, hospice, and palliative medicine, and has a PhD in clinical research. She is Chief of the Division of Critical Care Medicine in the Department of Anesthesiology, Perioperative, and Pain Medicine. Her interests really revolve around improving the delivery of effective and equitable palliative care, particularly for surgical patients and critically ill populations. Her research has been funded by PCORI, AHRQ, the National Palliative Care Research Center, as well as the Foundation for Anesthesia Education and Research. In addition to her clinical activity and research, Dr. Aslickson is also committed to teaching, coaching, and mentoring of young faculty. Our other panelist today is Dr. Michelle Gong, who is Chief of Critical Care Medicine and Chief of Pulmonary Medicine and Director of Critical Care Research at the Montefiore Medical Center. She is a professor in medicine and epidemiology and population health at the Albert Einstein College of Medicine. Dr. Gong received a degree in engineering at the University of Pennsylvania and then a medical degree at Yale University School of Medicine. She undertook training, postdoctoral training, at Beth Israel Hospital in medicine and as well as the Brigham and Women's Hospital in the Harvard Combined Program in Pulmonary and Critical Care Medicine. She has, in addition, a master's degree in clinical epidemiology from Harvard and has undergone the executive leadership in academic medicine fellowship training. Her expertise focuses on delivery of critical care medicine, particularly as it relates to management of acute respiratory failure and acute respiratory distress syndromes. She really has focused her research on improving outcomes of patients along the entire spectrum and continuum of critical illness, particularly the prediction and prevention of acute organ failure. She has been continuously funded for her research by AHRQ as well as NIH for over 20 years. She is currently focused on COVID-19 and ARDS, prevention of delirium, treatment of severe influenza, and apropos to today's discussion, big data and predictive analytics in terms of risk prediction and effective clinical decision support systems. So with that introduction, I would like to welcome our two discussants who will begin our conversation on the very interesting topic of the future of critical care. Thank you for that lovely introduction, Dr. Zimmerman. We are so excited and honored and grateful to be able to speak to you about this important topic today. So as already said, this thought leader session is on the future of critical care from artificial intelligence to Zoom family meetings with Drs. Gong and myself. We have no financial disclosures to make. We're not going to reference any unlabeled or unapproved uses of drugs or products. And the research support for both Dr. Gong and myself are listed here, and we are grateful to these organizations for supporting our research work. Also, I think we need to make a disclosure about predicting the future, because we're talking about the future of critical care here. And we actually have some wisdom from the past that can help us when we think about predicting the future. So as Mark Twain said, prediction is difficult, particularly when it involves the future. Abraham Lincoln also weighed in with the most reliable way to predict the future is to create it, which is a theme we're going to come back to. So when you want to think about the future, why not also talk to a science fiction writer? So here's Isaac Asimov. You don't need to predict the future. Just choose a future, a good future, a useful future, and make the kind of prediction that will alter human emotions and reactions in such a way that the future you predicted will be brought about. Better to make a good future than predict a bad one. And of course, you can go to Octavia Spencer, who essentially says the same thing more succinctly and beautifully. With all that you touch, you change. All that you change, changes you. So as we're talking about the future, we want to emphasize that this is not a reactive process. This is a active process. This is something we all have a choice, and we all will participate in. I'm a big fan of dance, and I love a beautiful pas de deux. So whether it's like the top left, the Alvin Ailey Dance Theater, or Taylor Stanley and Daniel Applebaum on the top right from the New York City Ballet, or the Paris Ballet in the bottom right, or the first all-female tango at the World Dance Competitions a few years ago, which is Liliana Chenlo and Yuko Artec. Everything we do with the future, it's a pas de deux, it's a dance. It's us in the future. We can impact the future. So we're going to talk about a lot of things that may be coming with the future, but we all have advocacy. We all have agency. We all have a role in enacting this. So if you don't like something that's coming or you disagree with it, be a part of it. Be one of the dancers. Get out there. Make a change in your local environment. Make a change nationally. We are leaders in this, and we want to emphasize that. So we also want to emphasize that this work is not the work of Heather, Michelle, and myself. It's the work of a wonderful, interprofessional, multidisciplinary task force of critical care providers that Society of Critical Care Medicine were so kind to help us interact with throughout this whole process as we go into discussing and thinking about the future. So on that note, I'm going to hand it over to Dr. Gong for the next part of the presentation. Thank you so much for that, Rebecca. So as we shift now to actually thinking about the future of critical illness and critical care medicine, it's important to understand, actually, the scope of critical medicine has really changed and advanced in the last few years and will continue to do so. We need to think beyond the walls of the ICU and even beyond the walls of the hospital to when critical illness may start with the patient. Next slide, please. And that means actually thinking about the reach of critical care medicine, even actually among the patient's home, but also just as importantly, next slide, when we actually try to send these patients back home, which is the goal of our critical care. Next slide. So we've always had an ICU without walls with rapid response team going outside the ICU. But now with emerging technology, we have the ability to do away with even the walls of the hospital and reach beyond to the patients to when they're home. If you click on the next slide, you'll see that actually what used to be thought of as something unattainable, diagnostics and treatments that start at home, has in the last couple of years become a reality. We used to send patients to get flu testing at urgent care centers or a primary care doctor. Now we have rapid antigen tests that can be done at home to diagnose COVID-19, all developed in the last couple of years. And if you're in a state like New York, if you have COVID-19, we now can deliver medication to you overnight so that you can start treatment at home and avoid hospitalization and severe illness and also avoid spreading COVID-19 to others in the healthcare system. Places like Amazon, as well as other places, have already started testing drones for faster delivery. So in a place like New York, you can have overnight delivery pretty easily. But in rural areas, these drones may be delivering medications to the patients at their home to where they need. Next slide. And it is not just these kind of diagnostics. We now actually have technology to remotely monitor a patient on a near continuous basis. All of what you see here, the watches, the EKG machines, pulse oximeter, we already use. It is available. In fact, patients can even purchase it. What's going to be different in the future is actually how these devices interact with the healthcare system. Next slide. As healthcare systems start adopting more and more of these technology with remote monitoring for both patients that haven't come to the hospital or when they leave the hospital, there will be more of a centralization where these data is going to be falling into almost like a tele-ICU mode for illnesses of patients outside a hospital system. And it may go to a team who will sort this out and contact the appropriate provider, but it also may be to automate a system that will source through the data and alert automatically based on an algorithm. Next slide. And it's not just these devices. We're growing to have an increasing understanding that some of the basic aspects of things that we take for granted can now be analyzed to help us identify patients who have a higher risk of developing disease, respiratory failure, and other conditions. So simple things like our concept in the past of like, well, how did the patient sound on the phone? Well, that can now be digitalized. So next slide. Having actually smartphones, smart devices in which a patient can communicate with the healthcare providers, the speech of the patients, the cadence, their breaths as they talk, how complex those sentence structures are, all falls into patterns that can now be analyzed to look to see actually its association with deterioration, worse outcomes, or depending respiratory failure. Next slide. This has been done mostly actually within the clinical research setting, but has been found validated for predictive mortality in patients with heart failure, respiratory distress from COVID-19, cognitive dementia, and developing cognition issues. As these devices and technology improve in the future, they can easily be incorporated into our assessments to be able to identify those patients who may be at a higher risk for additional testing or increased monitoring from other devices as well. Next slide. And then there's the issue of once you've decided that they need to come into the ICU. In urban settings, sometimes our ambulance will only be a few minutes away from a real resource hospital where everything you need is there. But a large number of populations still lives in areas in which it's not so accessible to a tertiary or quaternary care center. Next slide. But we're seeing actually these mobile units, and next slide, that has almost functioned like a mini ICU, an ICU room, with, as you can see here, in a mobile stroke unit, their own CT scanner. And this is actually already being in use in Australia, especially to reach those patients who are in more rural areas that takes a long time to get to a well-resourced hospital. And it's been shown to actually lead to an earlier time to thrombolytics for strokes and other interventions. Next slide. Next slide. Now, all of this technology and all of this monitoring is only going to lead to more data. And that's an issue because we're already right now have a lot of data. So if you can see from this figure, the amount of data for a per hour generated by a patient in the electronic medical record basically triples as they transition from the floor into the ICU. So there are over 1,000 data points a day for a patient in the ICU, not counting vital signs, which is checked almost every hour to every two hours. Next slide. This inevitably will lead to information overload and cognitive errors. Next slide. So whereas technology will be a feature of the future of critical care medicine and critical care medicine in general, it tends to be a data-rich environment, the revolution that's going to come and a future paradigm coming down the line is not going to be in data acquisition, but rather in how data is used and how it is used to help us do the work that we need to do to take care of the patient. Next slide. So artificial intelligence is already embedded in our everyday life. Every time I use a Waze to find the best route to my destination, or to use my camera and autofocus or Siri or get a call from my credit card or use Google, okay? And next slide. And not surprisingly, the use of artificial intelligence has been growing exponentially in healthcare. If you look at actually publications in medicine that uses artificial intelligence in one way, shape or form, or next slide, if you look at actually the market share in healthcare that's consumed by the artificial intelligence, there's been an exponential growth in these areas over the last 10 years or so. And much of this investment in artificial intelligence is actually done by private industries and vendors. Next slide. And it comes in different areas. There's the area of risk prediction. How do we use the rich data that we have to identify those patients who are at high risk of clinical deterioration, cardiac arrest, death, sepsis? And some of them, as I mentioned, the Groffman Index and even the EPIC sepsis prediction is done by vendors rather than necessarily within the healthcare system. Next slide. And then there's diagnosis. Next slide. Already, actually, many healthcare system has purchased software to help them improve their ability and their workflow. So radiology is a good example of that. Many health system has already purchased software to help them with tumor screening, with diagnosis of intracranial hemorrhage and stroke. Similarly, next slide, we see that in pathology as well to help with reading. In critical care, next slide, we're going to see also an advancement in terms of closing the loop and optimizing management, using data from the patient as well as their response to treatment to further tailor the treatment to create almost a closed loop system. We see this already in ventilators. Next slide. That have automated weaning and spontaneous breathing systems. In addition, next slide, we see it also in terms of drug tritation. For example, insulin dose adjustments for patients based upon their glucose. Next slide. Now, many of these things that I have discussed, potentially promising, but they're imperfect. Some of them are not very well validated and some of them performance still leaves a lot to be desired. So why is actually the artificial intelligence in healthcare not nearly as seamless and reliable as what we oftentimes use in our everyday care? Part of that is because of the volume of data available for validating these and improving these algorithms. On the internet, there's a vast amount of data, but in healthcare, we're oftentimes crippled by the fact that actually data tends to be siloed and can't easily be shared. That is changing. In the future, we'll see more of a move towards what's called a federated data. This is an example from the UK where they envision a system where research institutes, hospitals, general practitioners, the data all flow into a common platform that has the ability to anonymize, to provide security, and to link the data from these different sources and then also provide applications for data analysis that can drive care, quality, research, as well as actually public health. Next slide. This federated data system in healthcare has been advocated by the World Economic Forum. The next slide. As well as actually recognized to be an important aspect of research by the NIH, as well as actually something that's in the vision plan for Canada and their National Health Service. Next slide. Such a federated data system in critical care can enable a better utilization of personalized medicine where treatment is now tatered to the patient's response to the treatment, as well as to other data that we can leverage from their environment, from other sources. Next slide. It will be vital for surveillance to look for the next pandemic to understand where the resources are within. Next slide. In clinical trials, over the last few years, the most successful trials in COVID have been platform trials that basically leverage different interventions tested on a common platform that looks at the same data and the same outcomes. That's a perfect setup for a federated data. Right now, a lot of it is collected manually, but there's definitely movement already to making this automated by linking the EHR system. Next slide. Ultimately, what we're hoping for is that this will all accumulate into a learning healthcare system platform, where in the day-to-day life that we use to take care of patients, there is an influx of data. The data then gets analyzed. The results are fed back to the clinicians and to the healthcare system so that you can make improvements and test the next intervention. With that, I'm going to hand the discussion back to Rebecca, who's going to talk about other features that may be important in the future of critical care medicine. Thank you, Michelle. Just as Dr. Gong was talking about how technology will revolutionize how we care for our patients, it also will inform how we as critical care practitioners work and how we work smarter and better and potentially using technology to reduce burnout. To give a taste of this, this is a study that came out a few years ago from the Armstrong Institute for Patient Safety and Quality at the Johns Hopkins Hospital by Mike Rosen and his group, where they actually recorded over 356 work hours from 89 four-hour nursing shifts in an intensive care unit. They had wearables that the different nurses would wear as they were working around the ICU, as well as sensors so that they could detect where were the nurses, what were they doing, even to the point where was there a burstiness to their speaking? So were they talking and such? And they were able to actually designate which activities were associated with surveys that were done by these nurses for activities that increased burnout. So this type of technology can not only help us to take better care of our patients, but it can help us as ICU practitioners to work smarter and better and to even detect burnout or situations that are likely to enhance burnout. And switching over now a little bit, rather than thinking about technology, we want to think a little bit about the diverse ICU team. And yes, it's important now, and it's going to be even more important moving into the future. So whether it's nurses, physicians, dieticians, respiratory therapists, speech language therapists, occupational therapists, critical care pharmacists, physical therapists, rehab therapists, cognitive therapists, the care of our patients is getting more and more complex as their diseases are getting more and more complex. And we need this diverse team to be able to support the better patient outcomes that are possible. And if they're important now, they're going to be even more important as we move into the ICU of the future. And why is it important? Well, our patients are getting more and more diverse. Our people are moving around more technology that was in one place is moving so that it is moving into other places that may have before been considered low resource. There's a priority that there is equitable critical care, wherever you may live, be it rural, urban, whatever country. So that means all of us need to be able to take care of diverse patients. And on that note, we need to be diverse ourselves. So it's not just that we are able to compassionately and with cultural sensitivity care for our patients. We will do that better when we as an ICU team are diverse. So continued focus on that. Everything from racial and ethnic diversity to abilities and disabilities, you name it, we need to reflect it and to be aware of it. And why is it important? Well, here's just some data from the ethicist study where they were looking specifically at end of life care across different continents. And you can see the purple bars are patients who died in the ICU due to withholding treatments. The salmon covers colored are withdrawing treatments. The dark blue is actively shortening the dying process. The purple is failed CPR, or that would be a patient who dies with a full code status. And green is brain death. And the interesting thing here to look at is look at the diversity across the continents. And think of how many of our patients move between places. Think about your own practice and how often you're caring for people who may become from originally a different country than you, or maybe from a different culture. They may have different beliefs. So practicing culturally sensitive care means that we're able to recognize and support beliefs and practices, even if they may be a little bit different than what you would choose for yourself or what you would see as your culturally standard practice. So it's important that we have a diverse ICU team, even just looking at end of life care here, because it does vary between populations. And to be able to do that well, we need to communicate well. It's always been important. This has been a nascent area of research in the ICU for the past two decades, but it's getting more and more prominent with bigger and bigger studies supporting how do we better communicate in a culturally sensitive and compassionate fashion with our patients and their family members. So whether it's Doug White's trial here from the New England Journal of Medicine a few years ago that was about specifically a nurse communicator in the ICU to help the ICU team better communicate with families, or Amber Barnato's study about storytelling, or the whole plethora of Randy Curtis's work about communication in the ICU, we can study this and we can do it. So this was always important to ICU. And I think a lot of people used to think it was like an art, like, well, you know, some people are better communicators than others. More and more we recognize that it's teachable. It's teachable. You can teach good communication skills. I don't have any vested interest here, but Vital Talk is a very widely used education platform for communication for palliative care providers and trainees. There are many different ways to teach communication, and it's becoming more and more a part of how we train our palliative care workforce. And whether you're a physician, a nurse, a critical care nutritionist, or pharmacist, all of us interact with patients and families, and communication will be increasingly important across our multidisciplinary and interprofessional team, and it's teachable. And on that note, we're going to circle back to technology, because technology will also impact how we communicate. So we can talk about all the changes to telehealth that have been associated with the COVID pandemic and how we can use communication to not only help patients and families to communicate with each other, even when the family members may be distant from the ICU, but also, as well as communicating to our other critical care providers, so having specialists that are off-site, particularly for diseases that require a real subspecialty knowledge. But then also, and I'm very excited about this, how do we better communicate with our nonverbal patients, whether they're intubated or even delirious? How do we use technology in a more meaningful way to communicate with them? This is just some initial data of augmentative and alternative communication platforms for nonverbal individuals outside of the ICU, but that certainly can be moved into the ICU. And I'm incredibly excited about this, because I don't know about you, there's nothing more frustrating to me than having a patient who is trying to communicate to me with a breathing tube, and we are not able to. So I'm just, I couldn't be more excited and ready for technology to help us with this communication. And on that note, who better to help us communicate with patients than their family members? Family have always been important to the ICU patient and to our care of the ICU patient, and in the future, we will be integrating them more and more. So the Society of Critical Care Medicine already has had multiple guidelines for family centered care in the ICU. These are the 2017 guidelines. There's 2024 guidelines in work right now. And I've known to be clear too, when we're talking about family, this is a definition that SCCM has forwarded that family is defined by the patient, and it's just somebody that has a meaningful relationship with the patient. So it could be somebody who is familial related or not. Family is defined by the patient and their family. On that similar note too, the ICU in the future will look a little bit less like a hospital and more like a home setting. So less medicalization of the space, integration of devices so that it has a better feel for patients and families, less institutional, more home field. And then on a similar note too, the technology that we use in the ICU will get better and more integrated. This was my Flash Gordon moment when we were doing this on the task force of like, how, what, huh? ICU beds that take people's blood, that can measure vitals even better than what we already that does physical therapy. I mean, that sounded like something out of the Jetsons for me, and I did need my colleagues on the task force to say, Rebecca, this is coming. And this is work from Scott Halperin and his group. And you can see the other references that really are emphasizing how technology is changing even the equipment that we use in the ICU and in a better way. And as already hinted at by Dr. Gong, the ICU itself is changing and less in the hospital and more outside. So we already have ICUs in different resource settings, particularly in military settings that are mobile, that can be brought to where the patients are. This has certainly been used during the COVID pandemic when different ICUs have been built in public spaces to deal with COVID related surges. So we've already learned how to do this. And in the future, we'll be doing it more and more. So bringing it all home, really the future, the patient and their family are at the center of what we do. Yes, technology and data suffuses us already. It will continue to be the water in which we swim. But yet again, it doesn't change the fact that the patient and their family are at the center. They're the center of the environment we work in, the care team that we are and all the therapies we bring to them and how we use research, education and care delivery to do the best possible critical care for that patient and for that family. And to really talk a little bit at the end, we wanna talk about barriers to this future. So I think there's no quotation more meaningful to me than this one from Martin Luther King Jr. About that of all the forms of inequality, injustice in healthcare is one of the most shocking and inhuman. And I'm gonna hand it over to Dr. Gong to talk further about the barriers to this future and particularly that we need to make sure that this technology is implemented and delivered in an equitable and just way. Thank you. So in terms of actually the barriers to this future, there are multiple different barriers, but also understand that we have a role in being able to help address those barriers. So next slide. Okay. And next slide, we'll just put it all up. So, one of the big, big problems with a lot of the technology that I have mentioned and a lot of the promise of what Dr. Asterson had mentioned is that we don't actually know how well they work, especially in work in terms of changing the outcomes of patients, improving their care, and actually how well they work in different patient settings and different populations. And this is really important, partly because actually the more well-resourced centers may have the ability to detect the resources to be able to develop this, but in order for us to address ingrained biases and disparities, we need to make sure actually these things are tested on all different types of population and that well-resourced and less well-resourced medical centers have the ability to be able to utilize these interventions and these resources to look to see whether they improve outcomes or not. Next slide. So in order to be able to get to this future that we're seeing, we need to have a much, much better systematic and iterative evaluation of all the technology that we have. Currently right now, our testing of these and validation of these technologies and these algorithms are limited by small, retrospective, single center kind of a testing. It can't be. This needs to be much more widespread and it needs to be done over and over again in different patient populations. None of this can happen with the federated data unless there is enhanced security of health data and privacy. But I think we are going to get there just like we have been working on this on the public sector and the private sector, this will eventually come and there's already a lot of active work in this. Technology is a tool. So like anything else, the effectiveness of the tool will relate to how well we as the clinicians know how to use it and know how to communicate the results of it to our patients and to their families. So we need to be able to have a change in training of our providers so that they understand what this technology is, how to use it, how to disseminate it and how to actually communicate it, not only to the patients, but across the government, public health sectors, to other clinicians. And like I said, we can't just do this in rich, well-resourced institutions. We need to be able to have the ability to provide this kind of technology and these access and to include the data from poor resource institutions in order to be able to reduce disparity in our patient population. Next slide. And indeed, this is something that has been recognized in other avenues in terms of looking at technology. And this is one example of just the basic simple things that we take for granted, internet access and how it can actually worsen disparities by not providing access to it. Next slide. Ultimately, we all have a role in creating this future, and it's going to take a multidisciplinary group. We can't just leave this to the data scientists, the cognitive science, the industry and the bioinformaticists to be able to do this because we need to take a role as the clinicians, the clinical investigators on testing these, validating this, and making sure that it does what it's supposed to do, which is to help us take care of the patient. And we as the clinicians have to also be strong in terms of our involvement to make sure that it fits into the workflow of how we see patients, how we take care of patients and what the patients want to be able to do. And lastly, actually, implementation science is going to fall into this because just having the tools, and even if it's effective, doesn't mean it's going to be used in the right way. And that's going to be a goal. Ultimately, as you can see from this figure, if you go to the next slide, all of this only matters if, one more click, we keep the patient at the center of all of this because all of this technology, all of these algorithms, it's just a tool. But we as critical care physicians have a role in developing those tools and validating those tools in having a say on how those tools should be designed to help us take care of patients and keep them at the center of our goals and our efforts. So I'm going to transition this over back to Dr. Astikson because we have many people to thank. Sure. And I think too, when we're talking about the future, this photo is from the Critical Illness, Brain Dysfunction and Survivorship Center at Vanderbilt University, which is a interprofessional multidisciplinary critical care research center that will be research of the future. We can't be doing it in silos. We need to be doing things together and with institutes that really can appreciate the multidimensional aspect of critical illness and how researching it, you need to have that skillset and that wide stakeholder group to help you do, ask good questions and complete good research. So as already mentioned, we couldn't have done any of this work without the wonderful members of the task force who you can see all named and listed here. These are true leaders in their field and we couldn't be more grateful to have them go on this journey with us as we think about the future of critical illness. Of course, Drs. Gong and I especially want to thank our other co-chair, Heather Meissen from Emory University, who's also one of the co-chairs of this SCCM Congress. She has been absolutely influential in this work and in leading it. We also would love to acknowledge the input and the friendship and guidance of Greg Martin as the SCCM president who's worked with us and interacted and communicated with us throughout this whole process, as well as the amazing SCCM support team, particularly Diana Hughes, but also Kathy Vermach who supported our task force for the last almost a year and a half. And so on that note, we thank you for attending this presentation and for being a part of the wonderful 2022 SCCM Congress. And we will welcome a Q&A and discussion on these topics. ♪♪ Well, thanks very much for that presentation. I would say as a non-expert in this area, your remarks, both of you are exciting, hopeful for sure, definitely overwhelming and sometimes scary. So I'll start with that prelude. Michelle, can you tell me how federated data can happen in the United States? I can see this in China, for example, Canada, yes, in the United States, how is this actually gonna happen? Yeah, that's a great question. And one that actually people would oftentimes say that it is not possible in the United States. But I think that it is, and we already are moving in that direction. Now, albeit it's not as quick as some of us would like, but it can. So think of it in terms of scale, right? So just in terms of scale, if you look around at a lot of these medical centers already you're seeing they have started to create a federated data within their own medical system, which is no small deals because now medical systems are getting larger and larger with multiple different kinds of outpatient practices as well as their different hospitals. So 20 years ago, that wasn't the case, right? So even if you're affiliated with something else, they oftentimes have a different EHR system, forget the outpatient setting, there was like nothing that was related. But as actually EHR vendors became fewer and fewer and by hospital systems are purchasing these systems to basically cover the entire hospital systems, you're now seeing actually larger conglomerates of hospitals that already are having some degree of this microcosm of federated data within their system. Now the question is, well, how do you link these different systems together? Now, don't forget, it is not an unproven concept. The entire VA system is on one data set, okay? So we do have that ability and we do have that capacity. It's going to take a joining of minds to some degree and perhaps actually a government mandate to make sure that EHR vendors have this new standard that they need to abide by. This is a conversation that has been tiptoed around and vendors are saying, well, we're already doing it, so you don't need to make us do it. But I think eventually it's going to come to a state where if the hospitals wanna be able to get federal funding, there's gonna have to, just like the meaningful use, they're gonna have to meet certain standards in terms of interoperability, in terms of being able to share that data from one system to another, because if a patient goes to another system, they have every right to their data and to be able to do that. They're already making efforts to make that a part of the standards. So it's not gonna be tomorrow, but I do think actually it's an inevitability. The United States can't be the only country out of the modern world that won't do this. Just as a brief follow-up, do we get it, does our country get it that whoever does this is gonna have a huge advantage in terms of healthcare research and development going forward? I don't know. I think there are people in the government who do know this, but I also don't know that it is at the priority of what they think is needed, right? So a lot of part of what we described is an investment in infrastructure that will enable a lot of the other things. So conceptually, the government kind of gets it and understands it. But in terms of actually, whether or not we need to have that infrastructure, those mandates to be able to kind of make these things happen, that's still a work in progress, I think. Okay. Rebecca or Michelle, either one, but more and more I find myself when I'm on clinical service telling my fellows and residents, get out of the room where all the computers are and where all the data is, and just go and sit on the bed, put your hand on the foot. Is it warm? Does it have a dorsalis pedis pulse? Talk to the wife, talk to the parents. How does that mesh with all of this other stuff that we have to figure out how to keep track of? The real humanity of critical care, and it's huge, that's why we all went into this, I think, versus really the possibility of just things being so much better. I'm glad you bring this up, Jerry, and I couldn't agree with you more. This is a subject that's near and dear to my heart. And I think it's gonna get into a little bit of what Michelle spoke about, where the future is not gonna be acquisition of more technology, but better integration of the technology that we have. So if there is a pendulum that swings, and we started in a place where we had very little technology, but a lot of time with hands-on patients and families, I think we're on the far end of the swing right now where we're spending a lot of time with hands-on computers and hands-on technology. And that really the future is going to be better integration of that in a person-centered fashion so that the right information is provided to the clinician at the right time to be able to give the right care to that patient and to free us up from being tied to the computer so that we can really focus more on the patient who is in front of us and their family. Which gets into the idea that if you're spending less time on the computer, you need to know how to communicate with the patient and their family. And there are many things that technology can do to help us work more efficiently when it comes to data management and acquisition, but a computer still can't, I mean, you read the things about a computerized cat or a computerized partner or a companion for people, but I don't think we're gonna have a computerized ICU pharmacist or ICU nurse who's rolling up next to the patient going, I hope you feel well, it's gonna take us. This is the most human thing that we do as medical providers. And we're gonna need to be able to do that and do that well and even better than what we did before. And because patients are more complex and our team is more complex, all of us need to be able to do it, not just the one nurse who's really good with families, like it's gonna need to be the pharmacist, it's gonna need to be the therapist. We're all gonna be interacting more with patients and families and each other to be able to provide this care of the future, which is why we're gonna need to be emphasizing these communication techniques, because you know what? It's a little harder to communicate with colleagues than it is to just type on the computer. Right, thank you. For both of you, you may have noticed a couple of weeks ago in the New England Journal of Medicine, there was a report on a closed loop system for insulin delivery in kids with type 1 diabetes. These are kids like five to seven years of age. And it outperforms on the target glucose range, lower hemoglobin A1c. My question is two points. Every day I see another article about a driverless car in an accident. And then there's the doctors, not just the doctor, the providers at the bedside ego. How do we come to trust these systems? Yeah. No, it's something that actually is true, not only in healthcare, but also just in technology overall. But in healthcare, you know, I think we have the greater responsibility in terms of the art of medicine, right? Of how we do our clinical judgment. So there are a few things. One is nobody's gonna trust it unless these systems work. Right, you have to prove that they're effective. Okay, and that means actually like anything else that we do. So, you know, you described the insulin. Well, you know, the early stages of these algorithms for insulin and glucose control, they weren't great to begin with. But like any other technology, you study it, you revalidate it, you retest it, you improve upon it, and then it gets better and better. And these systems learn on how to make that even better and better. But even still though, right, it takes education to clinicians so that they can understand how it can help them. Not just that it can work, but how it can help them take care of the patient. You know, Rebecca talked about all this time spent on a computer, and you mentioned that. That is all true. But I would say that there are very few physicians who will say that that is why they went into medicine, that that's what they enjoy, right? But the fact is, is that that's what we have to do because of, you know, documentation reasons. And because actually, in order for me to figure out, you know, what's going on with the patient, I have to go through screen after screen of data on the EHR system to be able to, you know, figure out the picture. Well, this is where technology can help because you only have so much time and you want to use that time on the patients who really need it, who needs that extra attention, who needs to be able to need that, that which only you as a clinician can provide. So one of the things is, in addition to needing it to be work so that the clinicians are trusted, you need to be able to teach the clinician of how it's going to use to help them as a tool to take care of patients. Maybe in identifying those patients that need the extra attention, that needs those things that those algorithms cannot do, like going to the bedside, feeling their feet, you know, and palpating and seeing actually if they feel the PMI on a patient or is it like dampened, those simple things that mean a lot to the patient, okay, and that they can only get, but that allows them to kind of, as a tool to redirect their energy to those patients at the right time, at the right moment that needs that kind of a care. So it has a lot to do with, you know, I tell the story of my husband who has a PhD from MIT, who is dead set against the GPS. He thinks that everybody should be able to read a map and that I, and we refused to get a car for GPS that, and I have no sense of directions, but he uses Waze now, why? Because after we did it in a testing, he realizes actually Waze works and it has data that he doesn't have, like light changes and, you know, an accident that may have come down the road, you know, 10 minutes later that he doesn't know. So it's a matter of actually education, learning how to use it, trusting it, knowing actually where it fills in on helping you do what you need to do. And on a similar note, I would say it's going to require more out of us. And I laugh when you're talking about the spouse. My spouse is an early adopter. I'm not an early adopter of technology. I always kind of have to look at it a little while, but I think as critical care practitioners, we will all be obligated to maybe not be the first adopter and there will be a range in that, but that we need to be teaching ourselves and requiring it of ourselves to be critically evaluating technology as it comes out, to have the skillset to critically evaluate that technology, to adopt it when it is helping us and the evidence supports that it's helping us achieve our goals and to de-adopt it when it's not. You know, you can't just add to the plate, add to the plate, you also have to subtract. And I think all of us have been in critical care or many of us have been in critical care long enough to see practices that we think are evidence-informed and then the pendulum shifts in another direction or we get more data and it's not, and we need to de-adopt it. So we need to be agile on our feet with this technology, not just resist it because it's different or it's new. Evaluate it. Is it going to help you? Is it not? Have humility. I mean, I'm an anesthesiologist. I know a lot of anesthesiologists who feel threatened by these closed-loop technologies for anesthetic gases in the OR. I always want to hold myself to what's the most important thing, the best care to the patient and family. I also, I have the humility. There's some things the computer can do better, but I also know there's some things that it will never do better than me. And so there's always going to be a place for me. There's always going to be a place for us in the system. And especially if we are the ones who are able to juggle and evaluate that technology and implement it in a way that improves outcomes. Great. And I have no idea how I functioned before Google Maps. I have, I can't even imagine. All right, let me send you out on this one. This whole thing right now feels like 1970s and 1980s, HIV, total genome, human genome project, and genetics explodes. But just like what you two are talking about, that technology is far, far ahead of actual application and especially around ethical issues. So do you have any insight, ideas about how we avoid unintended consequences with all of this? Yeah, that's a crucial issue. So one thing with technology that has, especially with things like artificial intelligence that has escaped to some degree in medicine from our grasp is actually how these technology gets validated and tested because before they become available. I mean, you can have these algorithms be purchasable with very little data about how good it is. And that's partly on us that we didn't demand that they be put up to the same criterias that we would. We would never use a drug unless it's been tested for safety and for efficacy. We should have a similar kind of demand with regards to technology, with card artificial intelligence. And as part of that, okay, because artificial intelligence do look at big data, one additional aspect of testing for adverse events is to be able to actually test whether this data works equally well or how it differs from different patient populations and different socioeconomic and different types of hospital systems. So it is really important what you just said. And this is why Rebecca and I had talked about how patient is in the middle, but how the future is actually a part of what we create. So whether it be researchers who have a responsibility to test this in the same rigorous way as we do on a new intervention that I do on a clinical trial, okay, to the clinicians who demand that they wanna see data on how this actually improves outcome and show me that it works for my patients or my type of patients. And us in terms of like how we teach people to use it and how we demand that it be integrated within our workflow. It shouldn't be developed by the data scientists. It should be developed with us in mind, with the patient in mind in terms of how we take care of the patient. I've got two points with that, Jerry. One is we talked about the diverse critical care team. Big part of that are critical care ethicists and biomedical ethicists and clinical ethicists. They are a key part of how we do good critical care. There are wonderful ethicists within the Society of Critical Care Medicine who are leaders. So they are part of that diverse team because these are hard questions and difficult issues. And just as we need the expertise of our critical care nutritionists for some of the nutrition related issues of our patients, we need our ethicist colleagues to help with this. So that's the first point. The second point too is maybe it's, I was an evolutionary biologist in training when I was in college, and I still very much root in some of the work of Stephen Jay Gould from 20 years ago, but that the idea too, that evolution isn't gradual. It's a punctuated equilibrium. So you have bursts of change and then you'll have a period of stagnation where maybe things will change a little bit, maybe things won't, and then you'll have bursts again. And then you get into pasture with a quotation of, chance favors the prepared mind. We need to be ready for that. And we need to be thinking about it beforehand so that when we do have a whole new paradigm shifting technology, I'm thinking of like CRISPR technology for gene editing or even something like a drug like Gleevec for management of gastrointestinal tumors. I mean, these are game changing technologies. That was a period of rapid evolution. So we just, we need to be ready for it, see it when it's happening. And then just as Dr. Gong said, be very critical of how is this actually going to help my practice? What are the outcomes? Just because it's new and shiny, doesn't mean it's great. How is this actually changing what I'm going to do? And really requiring some of the research groups, show me the data in this population or that. And it's a bit of a slippery slope because I got to say, I get folks who are like, well, show me the data in post liver transplant patients, three months out who are in surgical ICUs. And I'm like, oh gosh, that's a little bit of a small, nobody's going to run a trial in that. But this is where we start looking within a critical mind and saying, what is the data that exists? How can it apply to my population? And really so much humility. What is the ultimate goal? Who is the center? The center of our universe are patients and their family members. And how is this actually improving what's happening to them? And then also being honest of, we can't do what we do if we're not taking care of ourselves. So how is what we're doing feasibly incorporated in our environments so that we can maintain our own wellness and be the critical care providers that our patients and families need us to be to be able to care for them. That's perfect. Do either of you have any other closing comments or insights about all of this that you want to discuss before we close? I just want to emphasize what we emphasized at the beginning. We all create the future. It's the patita. You are a member, you are an agent. All of us in critical care are leaders in our own institutions, as well as in our regions and nationally and beyond. This is an active process. If you don't like what's coming down the pike, then be a part of it, change it. Think about your own environment. It is that patita and we can all participate in it. Great. Michelle. And I would add, predicting the future is always tricky, right? None of us are soothsayers. None of us has a crystal ball. So the issue here is, I don't want to give the impression that these are easy solutions or these are inevitable things because they're not, there are barriers. We've talked about it. There are pitfalls left and right, but that is true of anything. Anytime you talk about the future, there are always going to be barriers. There are always going to be a pitfall. The question literally becomes, as Rebecca was saying, how do we want to address that? Do we want to be left behind or do we want, as critical care physicians, to take a central role in saying, if this is going to apply to our specialty in how we take care of patients, how we interact with patients, how do we want to see it happen? Well, thanks very much to both of you. This has been a great conversation, eye-opening at the very least. So I would like to thank our audience for tuning in with us today and participate with some of your questions in this conversation. And I look forward to seeing many of you next year at the SCCM Congress in New Orleans in person. Thank you very much. Thank you.

critical care

technology

data

communication

patient outcomes

diverse ICU team

challenges

validating technology

patient-centered care