July 2023 - Hospital at Home

- Welcome, everybody, to the July meeting of the ACEP Telehealth Section. We are super excited for this month's panel, which is all focused around hospital at home. My name is Rishi. I'm a member of the executive committee of the ACEP Telehealth Section. I have the honor of being the host of this section. I work clinic as an EM doc at Mount Sinai. I run our virtual urgent care, and I have a loving liaison like relationship with our hospital at home program. But I'm not the focus here. Our focus is on our amazing panelists, who I'm going to have introduced themselves. They all have a totally kind of different view from each other on this kind of quickly growing care model that we all need to get more and more familiar with. And we're going to hear about more and more, heard a lot about it in the popular press and in academic journals, and it's only going up. Michael, I see you first on my screen, so I'm going to pass the mic to you.

- Absolutely. Thanks for having me on here. So I'm Michael Nottidge. I am the National Medical Director for Contessa Health. My background is critical care, emergency medicine. And I am happy to say that I somehow survived the pandemic. And I'm still practicing, but I'm the national medical director for Contessa. Contessa, as some of you may know, is one of the pioneers of kind of working out these models at scale outside of the context of a grant or kind of a research funding. And so we've been doing operating since 2015 before the pandemic. So happy to kind of delve into the details of that and kind of where things are going moving forward. This is exciting. Thank you.

- Amazing. And that word that you mentioned, scale, I think, is going to be something that we talk about a lot. I think that there are a lot of programs that do single digit numbers here, many fewer that do double digits, and many, many fewer that do triple digits. And sort of the reason why is going to be something I want to explore kind of deeply with this group here. Pete, I see you next up. Can I hear a little bit from you?

- Yeah, thanks, Rishi. So, hi, everyone. Pete Chang. I'm a hospitalist invading an ACEP meeting. And so I've been practicing for about ten years, held various different roles at Tampa General Hospital, the first being the chief medical informatics officer. So lots of kind of IT work in the background and then transitioned into operations, built our clinical command center with GE Healthcare. So we have this really cool 8,000 square foot clinical command center that helps run the inpatient and ambulatory operation. And then now I'm kind of in this weird position called VP of healthcare design, where I can kind of just make up cool things that I want to do for the health system. Looking five to ten years down the road for us and report directly to the CEO, which definitely helps with some of those goals and achieving them. And one of them was to stand up at home program. So not just hospital at home, but we're looking at things well beyond the confines. I don't think we'll ever do ICU at home, but I think there's lots of space around hospital at home going down the acuity spectrum. So looking forward to sharing some of those experiences with you guys today. Thanks, Rishi.

- Amazing, thanks. And then last but very much, not least, Jared, pass the mic to you.

- Yeah, absolutely. Great to be with you all. So I still practice. I'm an emergency doc at Mass General in Boston and professor at Harvard Medical School. I spend about half of my time working as one of the leads of our Innovation Lab at Mass General, where we're trying to basically infuse technology in ways that allow us to move care up the quality trajectory, and hopefully in the more affordable category as well. Though that's not always possible. We've taken a particular interest in hospital at home and been in that space. We started a program back in about 2016. I've been working with Bruce Leff, who many of you know is kind of the father of this movement for the last decade. And we've done a lot of work, particularly both through our Mass General program and building that out to be kind of one of the first five or six that got permission from CMS to do this in the waiver to now really trying to build out a 200 bed hospital at home program across our system. We're obviously early in that journey. Our AEC is now in the 30s, and it needs to be in the 200s eventually, but that'll take a while. But my particular interest is around using technology to reach that scale, whether it's supply chain management, remote patient monitoring, et cetera. So good to be with you all.

- Amazing. And I love that you mentioned Bruce Leff, because I think any conversation around hospital at home is incomplete without that sort of amount of history, Jared. So I'm going to lean on you a little bit to explain to a newcomer to the space. What is hospital at home? What is it exactly?

- Yeah, I think at the end of the day, we've done a lot of work trying to decide what that actually looks like. I'm also, I serve as the medical advisor for the World Hospital at Home group. And different countries kind of think of it as different things. Sometimes it's like IV infusions right in the home, OPAT. And we kind of come out and said, that's not really hospital at home. Actually, hospital at home is simply anything that you do to an acute illness that requires hospital level care. Right? So think daily labs, daily clinician visits, and usually therapies, right? So IV diuresis, IV antibiotics, whatever that might be. And so that's essentially what it is. I will say that I also think of it as the highest quality acute care model. And you might think that is a little controversial, but I actually would say we have pretty strong evidence that it is the highest quality acute care model for the right risk stratified patients. So we obviously can't take care of ICU patients at home. I think, to Pete's point earlier, maybe a very low acuity ICU from a community hospital, though I think we still many years. Well, at least a few months, if not a few years from there. But anyone that's properly risk stratified, not needing the OR or the ICU with any degree of high risk, it is the highest quality acute care model.

- Yeah, that's a strong take, and I like it. I want to push you a little bit on sort of the types of patients that it's good for, that sort of seminal, original 2005 Bruce Leff paper took patients from a few pretty defined pathologies that they thought would be good for it, or at least to start. And correct me if I'm wrong, everybody, you're better than me. But I think it was COPD, CHF, pneumonia, and cellulitis, essentially. Is that the limit? Is that kind of what we do?

- Is that for everyone?

- That was starting, anybody answer, that was originally for Jared, because he made the strong statement around the highest quality acute care. Is that kind of where we're stuck still, or do you think that the model sort of goes above and beyond that and has a lot more potential than that?

- This will be a great discussion, but, you know, we started with what we call the five C's, which is, I think, is all the ones you mentioned, and maybe complicated UTI, if you didn't mention that one, or one of them. But we essentially started with the five C's, and we've built out since then. I kind of helped us think through, like, doing this for rhabdomyolysis, which is a great use case. Not quite as common as the five C's, but just a great use case. We've gone on and take care of a ton of other patients, and I think the real holy grail with this is to think of hospital at home as actually being applicable to every patient in the hospital, period. Except for, again, those that have any degree of significant risk of needing the OR, the ICU, or some other procedure that needs to be in the hospital. But I think at the end of the day, we should be using the most expensive hotel in town only as needed, right? And so most patients, they might need to spend two or three days in the hospital initially, I think we're all emergency docs or some sort of physician that takes care of sick patients, right? Sometimes you just need the tincture of time. So we're not saying that everyone can go straight into hospital at home, but there's two different models. There's hospital at home as a transfer model. So from the floor you say, oh, you're stable enough. It's been two or three days. We've maybe done that procedure that needed to be done. Instead of getting your four to six days in the brick and mortar, you're actually going to get half of that time in the home. And there's also kind of the substitute of model where you might come actually straight from your home, which CMS is not reimbursing yet, but we've had a good experience doing that for the right patients. Again, think of, like your heart failure patients that's 20 pounds up, but totally hemodynamically stable, and your cardiologist just can't get them on oral Lasix to where they need to be. Or substitute a model from the ED. The holy grail is really thinking about this for everyone, but particularly, I think, in the post surgical world, where we actually don't have a lot of evidence yet, but I'll let Pete and Michael chime in as well.

- Yeah, I think from our perspective, we've seen probably around 115 patients so far. So we're relatively young in our hospital at home journey, but that's spread amongst 23 different DRGs. So initially, when we started our program, we took the stance to limit to the five C's, and we quickly found out within the first month, wow, we just don't have the volume that we want to have. And I think we sort of threw pathways out the window. I think we all can kind of create those on the fly. A pathway approach to care is an interesting one. I think once you're meeting the standard of care and the length of stay metrics. The pathway doesn't matter a little bit. Every patient is going to take a little bit of a different journey and so we are rapidly expanding in the surgical space post spine patients. As a hospitalist, it's been interesting to me around all the great work that our surgical partners are doing around length of stay. I thought we'd have some opportunity in general surgery and bariatrics, but they do such an amazing job and get patients mobilized and at home in such a quick manner that I don't think there's a lot of opportunity for hospital at home. But we are a pretty large transplant center, so we actually are enrolling our first post liver patient today, which is a super exciting experience for us, doing titration of immunosuppressants and looking at steroids and using a Dexcom to manage blood sugars a little bit more closely than we would be able to do here in the hospital because we don't even use Dexcom here in the hospital. It's Accu-Chek's Q4. So those are some of the things that we can do. We, like Jared said, are rapidly approaching that post surgical space mainly from a capacity equation perspective, even if it saves a day or two of an inpatient stay. I think the key component is if you have the backfill to fill the bed, it's financially feasible to move that patient out of the acute care setting to home. But that's kind of been our approach and we're sort of taking things by a service line perspective. So we meet regularly. We have about 13 or 14 different service lines at TGH. Our CEO, I think leadership support is very important, has mandated each one of those service lines to work with our hospital at home team in figuring out clinical pathways and identifying an appropriate patient population, whether it be medical or post surgical, that would go into our program. And then, so we have kind of a process that we go through to vet out. Do we have the numbers? Is it worth spending the calories and developing kind of what the care protocol is going to be? What cross training do I have to do for my nursing team? So my nursing team, they're not transplant nurses. So guess what? For a week, they get shipped up to the transplant unit to follow some post op livers so that when we get those patients at home, we at least have some [Instructor] and outs and a little bit of experience to lean on in addition to the caregivers on those units, both from a nursing and a provider perspective. But it's a little bit of insight into how we look at growing volumes and diversifying the types of patients we're seeing at this.

- Okay, so, I guess there's so much to unpack here. And I think this is truly a fun discussion. But let me start by saying I agree with Jared. This is probably one of the higher quality options that we can offer people who we would otherwise put in the hospital. And I think that's a key part of this definition of hospital at home, is if your hospital at home program wasn't there, would these people otherwise be in the brick and mortar? And I think if you can define it that way, then you kind of know who you're taking. And so just focusing a little bit on the hospital at home aspect of what we do. We started with the initial 2005 paper, what I like to kind of fondly call the big four you mentioned. And then, of course, Jared mentioned the fifth one, which we then call the big five. And then as a proof of concept, I think between Bruce Leff's paper and the Mount Sinai papers, it was very clear that this works and it's a high quality alternative. But then, in terms of scale, exactly what Pete found is what most programs find is if you focus on those, the concept is great, but you can't scale it. And so you have to have this be able to operate in a business sense. And so one of the things we've generally said is we look at this as really meeting the patient's needs where they are, right? And so if we can meet their needs where they are in the least restrictive sense, as Jared kind of pointed out, we take them, right? And so we started with the, initially, the big five in concept. There's a little interesting history about surgical patients, but I can go into that later. But now we can take more than 100 different DRGs. Of course, we've had a little more runway than a lot of other programs. I was just looking at one of our health systems, which has, I think, done 1,250 patients so far in county. But there's a lot of learning here. Where we learn, we make sure that we have a learning system. We look at the patients we take, we look at the services we organize around them in their home, and we constantly fine tune it and measure and see how we're doing. So that's kind of the hospital at home space. In the post acute space, the possibilities are even more broad because your post acute needs aren't necessarily tied to your diagnosis, right? And so, again, when you figure out what the patient needs post acute, if you can organize those services in a reliable, safe way, and do it in a way that scales from a business perspective, you can actually reliably meet the needs of the patient and also make sure that the providers aren't kind of lashed over a barrel in order to get this, because I come from kind of a world of organizing workflows for docs, and especially during the pandemic, if you made something difficult for the docs, they just wouldn't do it. And so a big focus of mine is, I always use the term for sepsis. For everything I say, I'm looking for the easy button. So we want to make sure that this lines up is a truly quality offering for the patients, which I think has now been, over the last two and a half decades and across multiple countries, been proven and well established, right? But then also make sure that it makes sense from a physician perspective. And so I think one of the things that we have found to be contributory, and I'm digressing a bit, is the change management aspect for the clinicians. And then, of course, making sure that it makes sense from a scale perspective, which is the other kind of arm, to how we think about this.

- I love that you mentioned change management, because I think that's such a huge challenge for every organization that tries to do this, is trying to convince every single other person of the organization that they want to do this. But I want to stick to the very concrete and logistical, especially for folks that might not be as familiar with the program. Michael, I'm going to pose this to you because you've helped so many health systems get this off the ground. When I'm an ED doc, I'm a pretty simple guy. I admit someone from pneumonia, I know what bed they're going to. The transporter is going to take them there, the IV is going to stay the same, the pump is going to stay the same. I know the nurse is up there. I understand that. How do we make the home a viable care setting, and how do we get the resources that we need into the home in the first place in a really concrete, logistical way.

- So, I mean, the first thing is you really need to understand the driver in each part of the hospital and the same way these programs were put together. And maybe I'll step back a little bit. These programs, when we started them in 2015, the kind of path forward was figuring out all the elements that stood in the way of this progressing across America. Where it progressed in Australia and Canada, was you had reimbursement issues, you had physician and patient familiarity, you had EMR access. Where do people document? You had just the logistics and the cost of getting things done. And so we started by saying, okay, let's figure out what's important to payers. And so we started doing this from a value based perspective, because payers cared about the value and patients cared about the value. And then we went to say, well, what is a workflow? Because when they tried it initially, it made sense on paper, but the docs hated it. And so we sat down and said the hard lesson of if it doesn't work for your clinical team, it's not going to grow. You have to take care of the patient. You have to take care of the people who care for the patient. Right. And so the first thing we do, actually, is we have a pretty extensive discovery process where we sit down and ask ourselves, okay, conceptually, we know what you all do in the emergency department, but what do you do in this hospital in this emergency department? And how is this an advantage or a problem for you? How is it a problem for you now? And then we kind of use that to build a workflow that makes sense for them. And each hospital health system is a bit of a puzzle to solve, but they typically kind of lay out in the same way where we join their workflow and make sure that we're not having an ED doc get to the end of their workup and then saying, pause, now I'm going to start my process. That'll never work. From a very practical perspective, we should be able to, at the same time as they're making their disposition decision, work in parallel so that we can potentially siphon them out. The key to being able to integrate from the ED is have the right patient selection and have the right combination of processes that amalgamate the services so that we can get them out in a timely fashion. And the key is, I remember this old adage of two people are running from a bear and one is running faster, and the other person doesn't have to run faster than the bear, just has to run faster than you. And so in beating the hospital's innate admission time, the turnaround time in the ED, you can bring value to the system from that perspective. And then you have the ED docs on board with you. But if you come in and the ED Docs take four hours to get their patients out of the ED that are being admitted to the floor and you want to take eight hours, it's not going to work. So little practical tips like that matter, and then also figuring out where your transportation time is, figuring out all the elements of building the model together to make it work for the patient.

- Yeah, that makes a lot of sense. And I'm going to stick with you just for a second. Are the meds getting there in Ubers? Are the nurses taking the train? Is the patient's home bed just a hospital bed? Like, what does that actually look like?

- The answer is yes. So again, it depends. The first program we launched was in middle Wisconsin, central Wisconsin, if you're familiar with that. The second program we launched was in New York City. So in New York City, yes. Sometimes the nurses are riding the train because it's actually much, much more efficient than getting in a car many times. In central Wisconsin, they're in a car. And so sometimes it's courier, sometimes it's the nurses themselves driving directly. In general, I mean, Ubers are possibly an option, but in general, we haven't used those services because we need reliable services. And then some of the transportation is, we're talking about meds specifically, but some of the transportation is done by EMS. Typically, meds is a combination of either the patients taking them home, depending on when they're ready, the nurse taking them, or courier taking.

- We do something interesting in New York, and a lot of programs use nurses as sort of like the primary frontline sort of care delivery person. Up in Boston, Jared, you guys do things a little bit differently and you use medics. I'd love to kind of get some insight into that sort of decision and how it's different and what kind of learnings that you've had working with paramedics instead of nurses to deliver a lot of this care.

- Yeah, I will say we actually started with nurses, and nurses are still probably the core in home presence for us. But we have used medics quite a bit over the last two to three years. And I think ultimately most programs are moving towards a hybrid. Right, where you use both if you're at least allowed to. Not every place is yet, but we found that medics, we initially used medics that were from other third party entities. We didn't own our own medics. We didn't train our own medics. And frankly, those medics are great, but they're often also taking 911 calls. And so that has proven to be challenging because again, you need to build a highly reliable system. These are patients needing hospital level care. It's not like your Amazon where it's order entry to fulfillment, and oh, whoops, sorry, it's not going to arrive today. That's not acceptable. Right, in hospital at home. So what we've actually ended up moving towards is actually hiring our own medics and training them up specifically for hospital at home. And I'd say most of us that have done work in the space know that a lot of their skills are very well poised to do this work. I think they still need a little bit of training, just like an internal medicine doctor needs actually some training to do hospital at home as well. So they all need a little bit of training. There's actually a program that we talked about this at the World Hospital at Home Congress. There's, I think, first of its kind, Master's Program in Hospital at Home out of Barcelona. My colleague there, David Nicholas, I think is how you say his name, is running that with some other colleagues. But you can actually get a master's degree in a hospital at home. And they actually, it's for nurses, it's for docs, it's for administrators. It's a really nice phenomenon, but I think it speaks to the reality that everyone needs a little bit of training up in this space. But our paramedics have been great, whether we don't have enough nurses to deliver the care that day or it's an evening visit, when our nurses don't necessarily want to be doing the visits and our paramedics are more used to that time frame, we've often leveraged them in, whether it's those two main phenomenon, and they're giving IV medications, they're getting labs. Some of those are point of care labs, and some of those need to be brought back to Mass General to run the lab from our lab core. But we've had a really good experience with them. You can use a really powerful dynamic of a facilitated video visit where they're actually there, which I think probably a number of us had of experience with, but we've had a pretty positive experience.

- Yeah, I would just add, I mean, I think that's a really good point. We certainly are in support of a hybrid model where the state allows, because the nurses are nurses, and they're vital, but also the paramedics have an arsenal on hand. The heuristic thinking in hospital at home has to be different from the walls of the hospital, where the nurse can walk down the hall or walk to a Pyxis and grab the meds. There is a bit of a turnaround component for the average nurse where they're going to have to wait for the meds to get there, whereas depending on the symptom or the issue, usually the paramedic has some meds that they can apply immediately, so not like they're replaceable. But they're certainly complementary.

- Yeah. I don't know if any of my clinical leadership is on, but my schedule is free to go to Barcelona to take or take a couple of classes in a Master's in Hospital at Home. So feel free to send me. I'd love to do it. Mike Ross asked a really great question in the chat. Some especially diagnostic modalities kind of have to happen in the hospital or maybe somewhere in the community, whether that's imaging or procedures or anything else. Pete, how do you guys deal with this at Tampa? You think someone's a good candidate. They just need that one or two extra things that usually you could just wait for and admit someone, but now you really want to get them home. How do you do it?

- Yeah, so that's the beauty of kind of coming from the command center space. So I still have easy ways to fast track any pending exams prior to the patient leaving to home, if it's one of those, what we call, step down patients. So from inpatient to hospital at home, we do try to make sure that the ED patients, if they're coming from that route, that a provider from our team sees the patient into ED and makes the final approval that no additional imaging is needed. So try to wrap everything up before the patient leaves the facility. That's the easiest thing to do. If something develops within the home, actually, going back to medical school and our stethoscopes is the best diagnostic tool that we have. And doing a really solid physical examination, I put my APPs through a pretty rigorous physical examination course with me directly. So it's back to your roots in the black medical bag that physicians used to take out to homes. That's all they had. They didn't have CT scans and ultrasounds and MRIs. And so good heart, lung, abdominal examinations. We do a little bit of a dermatology, basic rash, mainly for medication reaction rashes that they may encounter when they're out there. So we've kind of developed some of that curriculum, which definitely helps. If a patient does need an imaging procedure, we actually use Lyft, and we'll take the patient to our outpatient imaging centers. Again, the goal is to decompress the inpatient environment, so it makes no sense to bring them back to the big house and try to do studies here. We've had, through our 100 or so, 120 or so patients, we've done one MRI, a handful of ultrasounds, one CT. So it's not something that happens frequently. The ultrasounds were around. Simple thing, right? Patient has bilateral hydro with urinary retention. Foley gets placed. There's a UTI. Send him home with IV antibiotics. Oh, by the way, what do we do with the Foley now? So we'll take them to an outpatient imaging center to just re image and see if we can try to do avoiding trial and remove the Foley. The next frontier for us is portable ultrasound and training the team. So that's another diagnostic tool that we can do a lot with. And I would probably argue if a patient does need an MRI, there's probably something else going on. They haven't declared themselves or there's something acute going on, and they would kind of exclude themselves from being a candidate for the hospital at home program. I think the MRI we did was just patient needed it for follow up care, and we just decided to do it on the last day of admission for them. There was a little bit difficulty in getting the patient to a facility for some mobility issues, so we bit the bullet and took them there. But I think imaging is a challenge. We have not explored portable imaging into the home yet. I know there are a couple of innovative companies that are doing some portable imaging. I tend to think the quality of those studies aren't the greatest at this time and really want to again focus in on some solid physical exam skills. It's an interesting one, but again, good neuro exams is another part of the training. I remember one day we opened up a remote ED, freestanding ED, and someone called in a stroke alert and spoke with the resident, and the attending said, "Well, did you do a neuro exam?" And she said, "No, the patient is getting a CT." And she's like, "No, I'm not going to do anything until you do appropriate neuro exam and call me back." So those are the type of things that I think we need to kind of reinforce, and it brings us back to the basics of medicine, which is exciting for me.

- Yeah, I kind of love that. If I remember that Bruce Leff paper correctly, it was amazing how many less interventions and studies that those patients got when they were in the home, and they ended up doing kind of exactly the same and were more satisfied. So I think it's a lesson that we can almost bring back to our normal hospital care to really ask ourselves, how necessary is this extra CAT scan or extra consult or extra whatever? How do you think about monitoring in the home and using technology? Jared, I know you're super interested in this, so maybe like a version 1.0. I'm just starting my program. What kind of monitoring do I need to kind of safely take care of someone in the home versus where you think things are going in the future.

- Yeah, great question. And we actually published on this in the New England Journal Catalyst. If any of you have seen this, I can put the link in the chat afterwards. But just our experience, because I think there's been a lot of work that have been done around people in process in hospital at home. The model has been around for quite a while. Really using kind of cutting edge technology is the key to scale, and I think monitoring, and really essentially we're docs that are used to seeing people occasionally crump and that makes us very uncomfortable with hospital at home when you're not familiar with this model of care. And so ultimately what we're all about is using technology to derisk that chance. And so remote monitoring in that space can actually do a lot of that. We're of the mindset that you don't need to be monitoring everyone, right, with some fancy bells and whistles technology, right? Your cellulitis patient probably is going to do just fine with just whatever you choose, Q8, Q12 vitals. Medicare allows for Q12 vitals if clinically indicated, at least in the current waiver. But for instance, and we published this as well, I can put this in the chat, in JMIR, that we've had really good experience, for instance, in heart failure around putting the Vital Connect device on patients, which gives us continuous single EDCG, actually gives us fall detection, respiratory rate, and even activity level. And we've had really good success over that, using it over the last five years, catching some folks that occasionally you're at your diuresing them and their K goes a little low or whatever it might be, and they go into a-fib with RVR, and you need to derisk that, right? And so we felt like, at least for our heart failure patients, the ACC guidelines say that's one indication, right? Acute decompensated heart failure is an indication for telemetry. We don't think it's needed on everyone, but we've had some good experiences being able to, again, rush either paramedic or a nurse of the home with a doctor, or APP by video, or if they're nearby, also can join and giving IV metop or IV dilt, and then having the comfort of just seeing them from our own homes, our offices, whatever, constantly, and seeing that heart rate come down. We know they're in A-fib per the rhythm strip. So we're really all about kind of matching the technology to the clinician, sorry, to the patient need, whether that's single lead telemetry, continuous O2, I think if you need continuous blood pressure, that's not you, right? You should be in the ICU. But there's a way to kind of think through this very carefully so that you really match what you need from it and just derisk the process. I will actually point out to Michael's point earlier about change management. We actually believe you should be doing remote patient monitoring for two reasons. One is for clinical risk management and the other one is for change management, right? There's actually a group of clinicians, whether they're your emergency docs or your cardiologists, or whoever it might be, that don't feel comfortable with this model early in its phase. And we're big proponents that actually doing more monitoring than you would have done in the brick and mortar is the right process there, right? The costs are going to be still way less. It is some cost to put technology on folks and have it monitored, but just giving them, again, the response time is longer. We should just be again de-risking the process, putting the right technology on the right patients. I will say just lastly, that the technology is on a journey. We all know that PPG is not perfect. I think almost every hospital in the US uses Masimo, right? It's great, but it's not perfect. Even if you're in the OR asleep, your anesthesiologist will tell you how even then, it's not perfect, right? And it's motion sensitive. So we're on a journey where we're working with a lot of technology companies to say, like, we need better products, but all of it's pretty good and actually allows you to do a lot of the type of monitoring you need for the right risk stratified patients.

- I will say I'm on two minds about that. On the one hand, I'm a bit of a device geek, and so the critical care part of my brain always wants more data. So I was geeking out about a lot of the technologies that kind of monitor activity, fall risk, continuous cardiac monitoring, some of the predictive models that kind of detect trajectories, downward trajectories, I love that stuff. On the other hand, we have several thousand patients across all models. And so when we look back at them, and one of my roles is to make sure, one, that the model is designed to give care, but also to make sure in retrospect, hey, looking back, were there things we could have done better? So I look at each case where there's an issue, and it's very rare in all the cases I've reviewed where I found that monitoring would have played a part. And so one of the things I always tell people is if I feel like they truly clinically need cardiac monitoring, that implies that I think they need a response time of a few minutes. And we all know right now in hospitals that there is a somewhat exuberant use of telemetry right now in general. But if they truly need it, then the hospital at home program is not for them. And so the first thing we do to derisk is we have a pretty rigorous process of vetting a patient's clinical risk, home risk. And the combined filter of that has actually been pretty good, where the percentage of patients where there's a need to rapidly escalate back or there's a complication is in a fraction of a percent, but you know, I was going to say, and I'm so glad Jared said it, is that there is a huge change management component to this and that is, you know, for a select group of patients, and I agree, probably there's a high risk tertile of patients in there, where having a monitor is both the reassurance to the clinicians and the family and the patient. And so even though right now, because of the filter we have, we've not really been able to demonstrate a benefit to our current patient population. There are two questions in my mind. One of them is, is there a population that could go if they had a monitor on clinically. I struggle with that, but I think heart failure is certainly one of those good candidate groups. But we take tons of heart failure patients already. So the other pieces is in a year's time, or in two years time, or in five years time, I think most hospital at home programs are going to be doing some form of remote patient monitoring, because that's kind of what's going to be table stakes to help grow the model. But I kind of constantly have some equipoise on the issue. It would be interesting to kind of maybe study it further. I really enjoyed your paper on this, by the way, Jared.

- Oh, thanks, Michael. And I think Michael's point is well taken. There is actually internal debate very much within the hospital at home community about who should be monitored, how much monitoring should be done. The Aussies, I will just say, have about 7% of their hospitalized patients in the home. 7%, right? So they're probably the most advanced of anyone in the world. But I will say they are flying me out there to speak as their keynote speaker in November because they feel like they've hit a ceiling, right? And so I think, again, technology is the way you get to scale. And for many hospital at home patients, you don't need a lot of monitoring. But as we move more and more up the acuity chain, again, thoughtfully and carefully, with better and better technology, it's how you get to significant scale. And so we're on a journey to try and figure out this carefully, but please reach out if that's an area that you have interest in. We'd love to collaborate.

- Yeah, Rishi, really quickly. This is an important topic, I think, in my brain, the jury is still out. We've taken the approach out of just the abundance of caution, as the tag has coined itself, to do continuous monitoring on every single patient. As we started off with small volume, it was just to wait and see. We've diagnosed four patients with obstructive sleep apnea and set them up with a remote monitor to diagnose and then set them up with treatment. We've had three patients go into A-fib with RVR at home that we would not have caught unless we were continuously monitoring. And our sample size is small. So I'm not sure if these were just coincidences or not. I think the points made earlier are, we decided to treat ourselves, and we use this as a tool to help sort of getting families on board with we're always watching kind of approach. Now that's back end infrastructure, too. It's not just devices, because you got to have somebody to watch it. And at the same point, I think that part of the program has been highly successful. The sensor that we use also does continuous blood pressure, FDA approved, which is really nice, because a lot of our patients struggle with blood pressure management. So to be able to titrate an antihypertensive and really see what happens after that patient takes their morning meds, what the response to the blood pressure is and how it changes throughout the day, I mean, we're really getting down into some finer details of the internal medicine brain, but we've been able to really streamline a lot of medication management as it pertains to a CHF patient that's having huge volume shifts from diuretics in addition to their goal directed medical therapy. And how do those two interact as far as them feeling good? Because they're actually perfusing what they need to perfuse, has been an interesting journey for us. So I think we're going to continue on where everyone's going to get monitoring. It's not a huge cost, it's not cost prohibitive for us. And so we're sort of just doing an experimentation and collecting data, and probably we'll add on to Jared's article and see what we can get in regards to, is it worthwhile and what is the cost benefit?

- Yeah, and it doesn't sound like there's a right answer here. And maybe, who knows? AI will be monitoring all of this for us in a couple of years, so you could be just ahead of the curve there. I want to make sure, there's a couple of great questions in the chat, but there's a piece here that we keep dancing around and haven't quite just really dug our teeth into. And that's the change management piece. And Pete, I want to start with you here because your program is sort of the freshest and you've had to do this very actively. I'm an ED doc. I might have to transfer my emerging cases to another hospital. I just got to the point in my life where I can just hit admit and that patient is completely out of my mind. How are you going to convince me that something more has to happen, that now they have to go home, that a whole bunch of other stuff has to happen?

- Yeah, well, I will have to say, I'm elated that this discussion is occurring amongst emergency physicians because I do think that that's the critical component to really scaling hospital at home. We've been lucky to have a pretty engaged medical director in our ED that has sort of been the champion. I think it's to your point. How does that message spread to the other attendings and residents and APPs that are working in the ED? It's a continuous change management push, banners in the ED is our newest thing, putting a care coordinator down there to continually reinforce and educate the providers. And it's not only from a physician perspective. We need everyone. We need the case managers. We need the bedside nurses who know their patients well in the emergency department to say, this is a great patient, right? A nurse is going to spend probably much more time with a medical emergency department patient than a provider will. And so in doing so, knowing the home dynamics and all of these other components that go into making a successful admission and enrollment into hospital at home, I think are key critical components. So we hit up the nursing forums, we hit up APP forums, we hit up the physician and medical directorships almost on a monthly cadence, like, hi, we're back at your monthly meeting. Let's have five minutes. And what we do is we actually bring back cases, because I think from a physician perspective, we also try to engage back, if somebody is a high referral to the hospital at home program, we actually send them a personalized note saying, "Thank you for your eight referrals in the last month. Four of them converted. And here's a synopsis of what actually happened to some of those patients." And I think the emergency physicians and the team actually appreciate some of that because a lot of the times those patients sort of get lost, and because it's a new thing, they actually remember those patients because they made the referral to us. So I think that's something that never ends. This is probably the hardest component of standing up a hospital at home program is the change management, not only with the emergency physicians. Hospitalist is a tough one because trying to unwind the ENM coding perspective in a private hospitalist environment like we have at TGH is an interesting one because you're stealing their patients. And so we've started to look at innovative models to let them still be the providers for the patient. And we provide the hospital service, so we're just kicking that off with APP leasing agreements for some of our private hospitalist groups. And then the last component is the consultants, and sometimes they're the most resistant. The hospitalist is on board for those step down patients, but the consultant doesn't want to let the patient go for a variety of reasons. So being able to let them know you can still video visit the patient, you can still charge for the visit, et cetera, I think is helpful in them understanding the role of how they can help and continue to manage patients. Our two biggest consulting groups have been, well, I should say three cardiology, pulmonary, and ID. And the infectious disease teams have been overwhelmingly supportive of taking patients home and completing IV, waiting on sensitivities, come back, switching antibiotic regimens if the patient is clinically doing well. Lots of really interesting things, but those were kind of our change management processes and I don't think it ever ends. You'll have new residents coming in right this month and the process starts all over again.

- Yeah, and I have to say, your guys' speed of kind of getting some buy in and growing your program has been super duper impressive. And the amount of private folks that you have there too, just a huge multi stakeholder problem that I really admire how you're solving. Jared, you guys are sort of on the other side, or at least in the middle of this process. You've been around for a while. The census is fairly high. It seems like penetration of the model is fairly deep at your institution. How does it feel being there? And does a lot of what Pete just said resonate with you?

- Yeah, I think absolutely. In fact, it reminds me of some of those early days when we were doing some of those things, and perhaps we need to go back to some of those things. Just to your point, Pete. I think, yeah, at the end of the day, I think also to Michael's point earlier, you've just got to make this the easiest thing to do. Like, it's got to be faster than admitting them upstairs. That's difficult to do in the early stages, but at least do your best to make it almost as fast as meeting admitting them upstairs. And our experience is we have had enough that most of our emergency docs, at least the attendings, are fairly comfortable with it. I think at the end of the day, though, most programs around the country and even the world have about a 5% to 10% escalation rate. And so I think, to Pete's point, I think that continued like, as best you can, at least if it's quarterly or monthly feedback, where actually this patient was escalated back to the hospital. Maybe you haven't finished your note and it's three days later and you notice, oh, my patient that I send to home hospital is actually now back in the brick and mortar. Man, I'm not going to refer to them again. That was not a good experience for me or my patient. I think it's understanding again that at the end of the day, patients heal best at home. So emergency physicians should think of this as like, I am giving the patient the best chance, if they're appropriately restratified, to heal quicker and better in the home environment, with the acknowledgement that about 5% to 10% of the time that good college try is not going to be successful. And recognize that having done now almost 1,500 patients just at Mass General alone and across our system, over double that. You have plenty of those experiences where the patient again needs to come back for whatever reason. Usually it's not because they're super sick, but they're sick enough that they need more careful monitoring or perhaps a different medication that you can't deliver in hospital at home. But at the end of the day, I think that change, man, it's still important to provide feedback to the emergency doc that, again, we're all in this together. We're trying to deliver the highest acuity, sorry, the highest quality acute care model, and occasionally it's not going to be the right model for that patient. And they might actually have to go back to the brick and mortar. And I think to Pete's point, providing some feedback to the emergency doc in the early stages is important in that regard, just to really level set on, we tried hard. It wasn't, in the end, the best for this patient, but they still did fine. Nothing serious happened. And that's kind of been our experience, right? All the way around.

- Totally hear you. Michael, I want to give you a chance here too.

- I mean, all these are really great points. I couldn't emphasize some of the things that Pete and Jared have said more strongly. One thing I will say, first of all, the feedback piece is one thing that I think has been an afterthought for people prior to the last two or three years. But it's crucial because there's oftentimes when a new concept and a new paradigm starts, it's very easy for it to get, quote unquote, bad internal press. And then that kind of feeds and heightens the anxieties that people have, like, oh, my God, you're sending patients home to die. So you have to be very mindful of that. But this change management itself is a whole field of specialization. And so being very intentional about engaging that methodology in the ED is important. We tend to start with just understanding what do the docs care about, what do the nurse practitioners care about, what do the nurses care about, what does the leadership care about? And then either craft a message that helps address that or change policy and incentives so that they care. Sometimes, actually, it's both. And so if you make it difficult, like I said before, it's not going to work. Because every time, I remember I was the, before I started running an ICU, I was the EDICU guy that would shuttle between departments, and they came to me and said, we have to get better at sepsis. And I said, if time was money and you turned all the docs and staff in the ED upside down, you couldn't get more. They're running around so busy, they don't have time to go to the bathroom. So you can't come with a solution that requires an extra step, right? And so I feel like change management clinically, and to focus on Emily's question for emergency physicians, I think it does require a unique understanding of the emergency department environment and what is really happening. And I say the same thing about hospitalists. And if you're going to come and ask hospitalists to do this, you have to understand what their day is like, and you have to design or redesign a system where for your program to grow, they're not killing themselves. And then you have to give them feedback about the things that went well, primarily. And then every once in a while, sprinkle in the things that could be done better. And you have to look for early wins and emphasize them early, because people need to know, just like patients, when they come to the hospital, they just want to get better and go home. Docs fundamentally want to know that their patients are getting good care and we're doing the right thing by them, and then also that they're not being killed in order to do just reasonable things. And so the first thing you have to do is emphasize to them, this is working. Jared said it very nicely. Just like you admit people to med-surg, and sometimes they have to escalate to tele, it doesn't mean it was the wrong thing to admit them to med-surg. I use the same analysis and logic when I talk to groups of docs. I say, "Listen, as a hospitalist, you admit someone to the floor, the next day you call me in the ICU and say, 'I need to transfer him.' I don't tell you that you messed up because you put them on the floor. I say that the condition changed." The question is, do we have robust enough processes, and does everybody trust those processes that we can identify if truly the condition changed or something could have been done better? And we're transparent and authentic about that. And so there's a whole lot of leadership, clinical operation strategy, change management, that goes into growing these programs. And a couple of years ago, before the pandemic, you'd start it and it would trickle up. Now we are trying to look for growth rates of 50, 25% in new programs, because we're trying to go from the crawl, walk, run, to people land, rapid adoption, preplanning, and then they launch pretty quickly. But that's very nice in theory. Each hospital is a unique puzzle. Plus, each hospital is dealing with, you know, the hospital itself has things they're looking at that they care about, and you have to be mindful of that. A hospital that is always at 95% occupancy is a very different hospital than hospital that is at 70%, 5% occupancy, and that trickles down across all departments, et cetera, et cetera, et cetera. And I will say one thing about consultants. Pete and I, we have the same thing. It's the same top three. But I've actually found in a lot of our health systems that our consultants sometimes are our best advocates, and we will sometimes approach them in the same intentional way, and it clicks immediately. If I talk to a cardiologist about outcomes for heart failure patients, they are the ones calling us and saying, "I have a patient on a Bumex drip, can you take?" And we say, "No." They say, "I'll call you back in a few minutes. Now, I have a patient on Bumex IV, BID. Can you take them?" And we say, "Yes." And so they are the ones, advocates. I will say the pulmonary has been half and half, which is interesting because I'm a critical doc, but some pulmonologist are just like, "What if I have to intubate the patient? Or what if patient needs to be intubated?" And there's like, no. So that has been a bit of a challenge, but it's fun. It's an interesting puzzle. Every day is a puzzle.

- I love it. Make it easy. And I love all your points about sometimes as an ED doc, you don't get to hear when that patient's doing great, whether they're back at home or whether they're up on the floor doing great. So really over communicating that and then showing them that escalation is okay sometimes. It's a natural part of clinical care. We all know that clinical courses take different paths, and it's okay sometimes, as long as someone's watching and something's happening. I want to move to a couple of the chat questions and then maybe leave, like even a minute or two at the end for open forum. Satta, we'll send you some of the health outcome papers, tons of data, tons of papers here. So we'll send you some of that. But I love your question on palliative care. Can someone speak to sort of palliative patients and their sort of utility in this program and whether it's better, whether it's worse, your experience with them?

- Yeah, happy to jump in. So the third arm of what we do is a robust palliative care at home program, and we do this within the joint venture structure and outside of the joint venture structure, so sometimes outside of a health system. One of the things I will say is, as we all know, palliative care is underutilized in clinical patients in general. And so we have deployed some of the usual palliative care triggers for our general hospital at home program patients. So it's part of my initial intake process to figure out what palliative care needs they have and loop them in. And so, apart from having a palliative care team, we usually try to leverage resources within a health system to say, "Hey, you know, how under or overutilized are your palliative care services? Can they be deployed here in a systematic way?" And we do that. That is actually, interestingly enough, so much more important for the post acute patients, which in retrospect makes a lot of sense. And so we have standard, we screen all our patients, standard for palliative care needs, and it's a palliative care automatic trigger. Get a consultation. And then we have just a more, so after patients have been admitted and let's say they had post acute needs and those are done, we have a standalone palliative care at home program which is targeted at patients who are not quite so sick that mortality is imminent within the next six months, but sick enough with chronic conditions where within six to twelve month time frame, they could be taken in. And we have a model kind of around making sure their quality of life is better, their choices are respected, and they have a lot of support for social determinants of health. And so I feel like, we started this program writ large early last year, end of 2021, early last year. And it's one of the things that most jurisdiction regions are very interested in, along with the post acute, because there's such a need in hospital and health systems. And as we know, in our patient populations, many, many patients end up getting a lot of things done to them that they honestly wouldn't really want if they had a chance to think through it.

- Yeah, Rishi, really quickly, we've used this as a vehicle for transitioning a patient to hospice care. A little bit different than the palliative care side of things, but it's been pretty successful in at least families having some extra resources during that time versus either transitioning to all by themselves at home or a hospice house, which is not maybe the best scenario for some patients and their families. It's a great opportunity. I think it's an untapped space that we need to take a little bit more advantage of for the betterment of the patients.

- Awesome. I think we're just about at time, so I want to thank, give a huge thanks to our panelists. They have a wealth of experience here and I know there are some questions around what if we're interested in starting this? Who do we ask? I hope it's okay if we have folks from the audience reach out if they have more questions for you guys, and we can send that out on the section chat. But I want to be respectful for everyone's time and let you all enjoy your afternoon. Thanks so much, and please reach out with any questions.

- Thanks, everyone.