September 2023 - Virtual Observation (Michael Ross)

- So welcome, everybody. As I had said in the section email, thanks to all of you who were on the last meeting. We didn't record it, which I think let us probably be a little bit more open about the challenges and great things we've been doing in telehealth. And also, I think it was just an amazing sort of survey of the telehealth and emergency medicine environment. So thanks for those who came and participated. Today, we are fortunate to have Dr. Mike Ross from Emory speaking to us in a real office. It's not a background. He's already told us that. To speak to us about virtual observation. And so I'm going to hand the mic over to you, Mike. If you wanna pull up your slides to make sure, and then I'm gonna go offscreen while you're presenting. So let's make sure your slides can come up, or I can bring up the slides from mine.

- Hey, can you see?

- Yep, I can see it. And if you wanna put it in slide mode. I'm seeing it where we can see the preview slides on the left-hand side.

- Slide.

- Slideshow.

- Slideshow. Is that okay?

- That's the presenter view, so that we can see your slide and the next slide, so.

- [Mike] Got it.

- Which is fine, just so you're aware of that.

- Yeah, so which is better, this mode? Oh, I don't see the slideshow. I guess I'm not used to doing that. Is this a better mode?

- That'll work, either one works fine.

- [Mike] Is that good?

- If you wanna, I think. Oh, there's no swap screen here, huh? Nevermind. Usually there's a swap screen if there's a double screen,

- I think it's fine. Maybe the other one you had, I think is fine, and just be aware that we can see the other slide, too, so.

- Okay.

- All right.

- We'll need a, just a second here. Yeah, looks good to me. I will just do the double screen mode, okay?

- Perfect. That works.

- Okay.

- All right.

- So put on your seatbelt. So, first of all, thank you so much for letting me share our experience with you. Such an honor to present to my telehealth colleagues in emergency medicine. And I'd like to cover, today's topic, obviously, is tele-observation, what I would call a sustainable win-win. And in my short time, we'll cover what it is, what works and what doesn't. And we'll review the model of tele-observation and the evidence, and the other questions surrounding that model. And then we'll end with dessert. We'll talk about disaster preparedness. You know, it's all novel things and interesting things. So let me start with what I call observation nomenclature. Now, observation is often confusing to people because it's both a verb, a noun and an adjective. And it's used in different contexts. It's been my career focus. So to clarify, there are observation patients, which I would call the six to 24 hour patients. These are people that need more than six hours of care in the ED, but less than 24 hours of care, if managed actively. They're a distinct population that's separate from an ED or inpatient or critical care. They predate all the payer policies, which is, and separately, we often think of observation services, which is really more of a payer definition, which is defined as the management to determine the need for inpatient admission. Now, interestingly, observation patients represent between 20 and 35% of people staying in hospitals. So they're a sizable part of population of patients in hospitals. And then, finally, and most pertinent, is observation settings where those patients are managed, which is defined by two variables. One is the use of protocols, and two is the use of a dedicated unit. And this will all tie into telehealth in a few minutes here, but I just wanna make sure that you understand. So a type one setting is a protocol-driven observation unit. And at the other extreme, no protocol, no unit, Wild West, everything goes. That's a type four setting. Interestingly, 2/3 of hospitals in the US do not have a type one unit or do not have an observation unit. So the type four setting is the most common setting in the US. There's a very large body of evidence supporting better outcomes in type one settings. Over the last couple decades, there have been nine prospective randomized controlled trials looking at conditions such as syncope, chest pain, AFib, TIA, asthma, all showing outcomes that favored care in the type one setting, lower admit rate, lower length of stay, lower cost, better clinical outcomes, less diagnostic uncertainty, et cetera, et cetera. So to kinda help get your head around this, allow me to tell the tale of three hospitals, okay? And hospital one is, you may be familiar with it, it's a hospital where observation patients are admitted to a bed anywhere in the hospital. They're indicated on this picture by the blue beds scattered throughout the hospital, okay? Now, and I'll use representative numbers here. These are not far off. The length of stay will probably be between 33 and 44 hours, their average length of stay. The total drug cost per case, not charges, but cost per case will be about $1,978 per case. Now, if you take those patients and cohort them to an inpatient unit and take care of those patients by protocol, you may bring the length of stay down to, say, 28 hours and bring the cost down to about $1,800. And through that efficiency, you may open, effectively, open inpatient beds, which are indicated by the pink beds in this slide here. You with me? Okay, so if you go on to, you can take it a step further and move those patients completely out of inpatient beds because, after all, these are outpatients, they're not inpatients. You have outpatients occupying inpatient beds. So if you move them out of the hospital completely, into an ED observation unit, their length of stay will come down to, say, 17 hours. Their cost per case will come down to about $1,342. And it'll be a larger net benefit. Now, if you kind of lay these three scenarios out, okay, and simply take, you know, an average hospital with, say, 3,000 observation visits per year, which is reasonable representative, and run the math out, which I've done for you here. If the difference between hospital A and hospital C are a type one and type four unit, the cost savings of that approach to care is about $1.9 million per year, okay? Now, the bed day savings. Now, a bed day now is if you take the length of stay divided by 24, that's a bed day. So total up hours divided by 24. So the number of bed days saved between the type A and type C hospital is 4,125 bed days made available. Now, if you take the average contribution margin of a bed day, which, if you back bill with a new patient, new revenue, it's about $1,000 per day, the contribution margin is a thousand times. That'd be about $4 1/4 million. The net impact, two very different things, cost reduction and revenue enhancement, that impact is about $6 million per year for this average hospital with 3,000 visits per year. Now, you may say, "God, that's a purely rhetorical calculation," but let me bring it back to reality here. Let me share our experience. At Emory Healthcare, we have six type one units that have developed over the last couple decades. So let's take the four legacy hospitals. And same thing, I kind of ran the numbers. CDU is our type one setting. It's a clinical decision unit. NOU is a non-observation unit setting. It's an inpatient bed. And patients are, you see the average length of stay and cost per case. And if you run the math out, on an annual basis, these four units saved Emory Healthcare $24.8 million just in cost reduction. If you add in the bed days made available, it made 10,526 bed days available. Now, is this a onetime thing? No, no. I mean, If you take all Emory Healthcare combine the CDU or the type one unit patients and the NOU, the inpatient observation patients, and track the census over the year, the top left graph is a comparison of census by setting. The blue line is the monthly CDU census. The orange line is the monthly NOU census. And you can see that we're managing the majority of observation patients, about 2/3. The length of stay difference month to month doesn't really vary that much. The NOU length of stay is trending up a bit, but over the last year, the difference is about 20, 24 hours per case. And then the cost per case also similarly differs on a month-to-month basis by setting. And, by the way, I won't bore you with the details. This is a case mix of the things that we observe in our type one units with the respective ED length of stays, CD length of stays. Average, 18 hours. Admit rate's about 19%. But this is not about observation. This is about telemedicine, right? So, and for hospitals, it's great. I mean, we decreased cost, decreased length of stay, lowered admit rates, improved inpatient bed availability. For physicians, hmm, not quite feel great. And it's really largely driven by CPT. CPT, which defines how you can bill for E and M services. And by CPT, a physician can't bill, even though we're providing two distinct and separate services, you can't bill for the two services. You can't bill for the ED visit and the observation visit. So unless you have a separate, unless you create a separate service, a separate group, if you will, defined by a tax ID. So let me explain. So we published this, Chris Baugh and others and I, right, a couple years ago, where we kind of ran out a different, we did modeling. We did what's called Monte Carlo simulation modeling based on the two billing models. Our data source was existing literature, national survey data and payer data. And what we found is that with the single service model, no matter how many beds you had, no matter what your census, there's net negative cash flow. So usually those units are subsidized by the hospital because of the tremendous financial benefit to the hospital. So that group from A, oh, hold by a subsidy from the hospital. With the two-service model, where you're able to bill for both, you need a critical threshold of 20 beds, as indicating with that tipping point, to reach net positive cash flow. So there's a 20-bed threshold to make the service sustainable, okay? Now, pondering this, several years ago, it occurred to me that I could actually combine two of our CDUs and come up with that magical 20-bed threshold by using telemedicine. So I started to dabble in that space and I realized that we could actually develop this as a service that we could share with other hospitals and expand our footprint, if you will. Interestingly, the CPTs, the CPT manual, the chapter on E and M codes, which really, E and M codes are cognitive services, things that you do with your head, not things that you do with your hand. In that space, telemedicine's been studied and proven effective for clinic visit, E and M code, critical care, E and M code, emergency visits, E and M codes. Really, essentially, nothing has been done in looking at telemedicine in observation services. So I realized that there's an opportunity for me to dip my toe in the water, which we did. Now, then along the way, this little thing happened called the COVID pandemic. Anybody hear of it? It's a bit of a game-changer. Because of my experience in this space, I was asked to launch telemedicine for the entire department of emergency medicine. And I guess I look at it as, telemedicine, as a tool that we manage, that we used to manage our time-sensitive conditions. And it covers the entire spectrum of acute care. So we're currently doing tele-EMS. We've done over 70 cases, where we're managing critically ill patients to rural South Georgia that have one-hour transport time to the nearest hospital. We're involved in SRDRS, which is disaster preparedness and response. We've just started a nurse call line, which is our first foray into direct to consumer. We've got, approaching 1,000 calls in a very short period of time. But today, we'll talk about tele-CD. We've also dabbled in triage, ERTD, in various triage models. But for the sake of staying focused, let me focus on the CDU part. So over the last couple decades, as mentioned, we've grown six type-one units. Here's 2021, '22 average metrics. I realized that I can combine the CDU beds of Midtown and Emory University Hospital. That would give me 29 beds or 8,713 visits per year. We've been staffing those two CDs for three years now. We have a very robust experience. We're in the process of expanding to two more hospitals within our system. That would have a combined census of 22 beds, 7,295 visits per year. So this is a sustainable service that we've found to be very effective. Now let me show how it works, and you probably can imagine. Previously when you did a CDU shift, you'd come in, you'd log in the computer with the APP and you'd run the list. You'd go through all the patients in the observation unit, review the chart, and then you'd walk room to room and round with the patient and make decisions. And then you'd move on, and at the end, you'd go to the ED. By combining, at the end, you stay on and this is your shift for the day and you co-manage the patients. Usually we'll pick up eight patients. You know, call admissions, call consultants, write discharge paperwork, do summaries, et cetera. So we've been doing this, but we kind of paused and said, "Is what we're doing right? I mean, is it worse, better or the same? Is it non-inferior to rounding a person?" So we did a non-inferiority, kind of the diff-in-diff retrospective, non-inferiority analysis, where we looked at 20,861 observation unit patients. We compared it to 23,000 NOU, or we had them kind of as a control group. Because it spanned the pandemic, it was really tricky because we didn't know if what we were seeing was because of the pandemic or not. So we had three control groups. We had the before period, we had one of our hospitals, Emory St. Joe's, that did not adopt telemedicine, but they used the same protocols. They model everything. So they sort of control. And then we used the difference between CDU and the inpatient observation patients to see if things changed. And to cut the case, we found that after controlling for all the confounders that we had available, the adjusted length of stay showed no difference. The adjusted admit rate showed no difference. The adjusted total cost, no difference. In adverse event, the adverse event rate showed no difference. So we concluded that it was not inferior, that rounding virtually in a protocol-driven obs unit was not inferior to rounding in person. Now, this was published in "Annals of Emergency Medicine" this year. You can download the article and kinda dig through the details as much as you wish, but, really, it answered one question. But in creating this virtual observation model, I realized that it's a different model of care, different model of care. So we're in the process of writing up what I would call the superior. We've done that non-inferiority study. Now we need to do the superiority study. In other words, compare virtual rounding with care in a traditional inpatient bed. And I can just kinda share two consecutive years of virtual rounding compared with NOU or inpatient care. 31,000 total visits, about little over 1/2 in the CDU or in the type one unit. The difference in cost per case for discharge patients is $1,882. Difference in of length stay per case is 19.7 hours. If you annualize or if you total up the total of savings over two years, it's $52.1 million. The bed days saved is 17,547 bed days. Oh, between the two hospitals over two years, though. Stay tuned on that. So this begs a few good questions. One is, you know, what if the payers come along and stop paying for telemedicine? Can this be used for hospital at home? What about virtual obs units? And what about this new CMS thing called a rural emergency hospital? Now, we, you know, we thought our biggest threat was if when the pandemic ends, that they'll say we have have to go back. And I realized that no, really, we have an APP onsite. We're rounding with the APP. The worst-case scenario, if they say you have to physically be present to staff the APP, I reached out to our local Medicare financial intermediary and asked, "Can we bill under the APP?" And they said yes. And to paraphrase, they said, "Yes, you can," much the same as an APP might see a laceration in the ED or a belly pain in the ED and see that patient independently or not. Now, we would still staff the APP just for our own internal quality utilization and other reasons. But, so we are, on some level, protected by that. There's a lot more nuances to telemedicine and observation billing I won't get into right now, but we feel somewhat safe with that as a backup worst-case scenario plan. What about acute care hospital at home? This new thing that I'm sure most of you have heard of. Can that be used for observation patients? Well, this is directly from Medicare's website. They say that yes, it can be used for observation patients. And rounding, on the physician, the physician has to see the patient at least daily, and this can be done remotely if appropriate. So it appears that hospital at home can be, if you have that developed at your hospital, can be used for observation patients. Well, what about this really new kid on the block, totally cool, called virtual observation units? We're fortunate to have the world expert in this topic on the call, Dr. Hayden. Thank you, Dr. Hayden. And these are two of her papers published, I believe, this year, where she looked at the perception and experience and impact of virtual observation at home, where the patient's observed at home and you're managing them remotely, with providers onsite as needed. And the provider perception is that they feel that they're heard. They have a better sense of meaningfulness, positive attitude towards virtual care. And patients, that majority would recommend this to others. They strongly have strong positive perception of courtesy, respect, engagement in care. So there's a lotta good preliminary data, if you will, in this area. As I said, I think it's the next thing. Now, we also kind of wondered, is this gonna replace observation in the hospital, in our type one units? And to answer that, I took the last two months' data for five of our Emory Healthcare CDUs. This is a rank-ordered list of our conditions with our metrics. And I simply broke them very broadly into things that I thought probably could be managed by observation at home. That's the green shade. Probably couldn't be managed by observation at home, that's the pink or red shade. And then, you know, maybe, maybe not, depending on what you develop. And, by the way, I can think of exceptions. This is like a throw mud on the wall to see what sticks kind of analysis. But, you know, from my estimate, you know, probably 20% could, and maybe another 1/2, 10, 20% could, but probably 60% might not. And, by the way, you know, that's assuming, that's of the total group. There may be people that might qualify, but they choose not to or other things. But I think the take-home lesson is, I mean, the denominator here is 3,000 patients. There's definitely substrate. There's definitely potential to develop hospital at home for a sizable part of people that we manage in our observation units. So I'm really excited to see where this goes. And in a lotta ways, I think the work that we've done has shown that virtual rounding is effective in a hospital setting. I think the next step logically is to say that model of care could be applied in the home setting. And I think that's what Emory is doing very effectively. So, let's see. And I guess the last unanswered question is this new thing for Medicare called rural emergency hospitals. And that's where basically they've taken critical access hospitals that otherwise might have closed and said, "Tell you what. Stop taking care of inpatients. Turn your hospital into an ED with an obs unit, where the people can't be expected to be there for more than 24 hours. Admitted patients would be transferred." I think there's a lot of potential to help staff these satellite rural ED, rural emergency hospitals, much the same as we have with our observation units. So onto the last. What about telemedicine disasters? Well, let me first say, you know, as mentioned, observation units have been my career hobby. Just as the emergency department is the safety net of the healthcare system, observation units are the safety net of the emergency department. And from my direct experience or, you know, knowledge of what others have experienced, I'm aware of six cases where observation units have been leveraged that way: one epidemic, one pandemic, three inhalation disasters, one environmental disaster. And in all cases, the observation unit played an instrumental role in helping the hospital and the emergency department adapt to the disaster at hand. So, which begs the question, you know, if your hospital has a regular daily use of telemedicine for an observation unit, it's better prepared to deploy that equipment for a disaster. In other words, if you buy like a cart and stick it in a closet until there's a disaster, guess what? When the disaster comes, you'll dust it off and say, "How does this thing work?" But if you're using it daily, you're more likely to be able to leverage that in the setting of the disaster. I think of a disaster as a bad day in the ER. I mean, really, a disaster is where the system is overwhelmed. In any disaster, there's gonna be a population of patients that are observation eligible, six to 24 hour eligible. We're currently developing observation protocols for disaster patients, chemical, biological, radiation, natural explosive, that can be managed onsite in a type one unit. And having telemedicine for your obs unit in a disaster allows you to very quickly bring in subject-matter experts for things that you don't usually treat, like radiation, chemical or biological disasters. It also provides flexible provider staffing pool for the sudden surge in patient volumes that you might be able to leverage on a local, regional and state level for the disaster. And it helps you to preserve their resources, such as inpatient beds or PPE equipment, because that's usually the, I mean, that's really what defines a disaster, is that resources become overwhelmed. So in summary, the type one units have become routine at our shop. We found it to be non-inferior, and we believe it's gonna be shown to be superior to traditional inpatient care. It's improved staffing flexibility. I can tell you, our docs would not go back to in-person. They love it. And when the pandemic hit and people were COVID positive and asymptomatic, we could put them into this position and have them continue working, if they felt that they were well enough to do so. So, and it can be leveraged in a disaster. So with that, I will stop talking and open the mic for questions.

- Great. Thanks, Mike. We really appreciate this. And it's just, I think it's so great for, as the expert in observation care and all the work that you've done beforehand, to also demonstrate the value of emergency medicine and observation care, to see you now in this telehealth space and continuing to innovate and continuing to show emergency medicine, creating sort of efficient and, you know, like you said, non-inferior, but also superior care. And so really appreciate you sharing this with us. And I'm sure that there's probably questions here. I know I learned, despite my virtual observation unit experience, there was definitely, I was scribbling notes with the beginning sort of primer on the observation care. So questions for anybody? I know some people have come onto screen, or you can always raise your hand or just speak up.

- I'll comment. John wrote a great point. We had a train derailment, and our CDU was key for how we managed flow. Right now, we're reaching out and asking for people's experiences with how they leveraged an observation unit in a disaster. We're writing a concepts paper on how obs units might be leveraged. So we're really hungry for any experience that anybody might have out there. So thanks, John. We'll take your name down,

- Carly, feel free to unmute and go ahead and ask your question.

- I'm just kinda curious, in terms of transitioning, you said you had someone coming in, working with the PA and then going down to the ER. That's kind of our current model. When you transitioned to virtual only, how did that impact kind of your staffing? Were they able to do multiple facilities? Were they still available? You know, was it a one-to-one ratio or did they have to kind of adjust how much time they had?

- Yeah, no, actually, the first thing I did to convince to myself that I wasn't compromising patient care, I, myself and my then-administrative fellow, we built our equipment and we rounded virtually. And then I came into the unit and rounded in person to see what I was missing. And I realized a couple minor things I was missing, like an axillary abscess or something, the APP was there at the bedside and they could examine. Now our equipment's so good, I don't think I'd even miss that. But, so once we did that, yeah, we don't go down to the ED. When we're doing our, it's Emory Healthcare telehealth shift, all that we do for that entire shift is round on the obs unit. We cover, I round from 7 to 8:30 at Emory University Hospital, 13 beds. Then I round from 8:30 to 10:30 at Emory Midtown Hospital, which is 16 beds. And we round at a clip. Actually, and before I round, I'll check in to say, "Hey, is there somebody who needs to be seen right now at the second hospital?" And once I do that, then we just, we rounded all the patients. But then when we're done, I roll up my sleeves and I help the APP. I'll take four patients from each hospital and I'll take care of 'em. If there's something I need their help with onsite, I'll call 'em, but that's like, it's probably 10% of the cases. I'll call admissions, I'll write discharge instructions that they'll print off onsite. I can write prescriptions that they can print off onsite. I, you know, can write the notes. It's a full shift. It's also, it's a nice change for our docs because, you know, you can actually, a lot of 'em can work from home. At first, I discouraged that and had people work from a control room at the hospital. And then with COVID-positive people, we started working from home, and I realized you can do this from home. And like it's the first time people are actually eating lunch in the middle of a shift. Like, wow. So it works really well. And all the docs that work the ED are the same docs that work the shifts, so. If they send something that's a mess, they'll hear it from their colleague, but that doesn't happen. Yes.

- Yes, just curious about one thing. You just said that you rounded at two different hospitals for this, and I take it they're probably major hospitals. Have you developed any protocols for any rural extensions that would still fit within the nurse practitioner, the NPI systems that you've developed?

- We have, by the way, if you're interested in our CDU manual, just Google Emory CDU manual and it'll come up. It's like it's free internet. It's like a 64-page document. We have all of our protocols. You can see what I'm talking, in granular detail, you can see. And we designed the protocol so it's not necessarily hospital-specific, so that if a doc lands at any hospital, it's gonna be the same care. Now, with the provision like one hospital has stress PET imaging, the other one has, you know, there little nuances like that. But really, especially for the treatment protocols, I think that's generalizable to almost any hospital. And, by the way, the protocols that we develop are for our health system. I can't say that they will necessarily work at yours, but our protocols have been used countless times for other health systems as a model to help them develop their own protocols. So you could take those protocols and adapt them to the local practices and resources available at your health system. And what's nice about telemedicine, I mean, this is the hugest thing, is, yes, you need about 20 beds, but you could cobble together. Like, you could do three obs units. You could do, I mean, I think 20 is the threshold that was mentioned in the paper, and that's based on assumptions of cost and staffing and reimbursement. But there's gonna be some magic threshold of which you break even or do well. You might have four small obs units that you could staff instead of one or two.

- Yeah, and I'll jump in on that one, too, extending on that one of what you're saying, Mike, on the virtual observation side, where it's the virtual obs patients at home that we've been working with. It's similar, where to be able to have, I mean, we haven't modeled it out, per se, right now, but to be able to scale enough to have a sustainable program, that we've been working with just our emergency department at Mass General. However, there's others in our system, multiple EDs in our system. And so it's that trying to fill our beds, right? We have funding for a certain number of beds, or we had funding for a certain number of beds, and you wanna fill them. And to fill those just with one emergency department is hard. But to be able to say like it probably is gonna be, to fill those 10 beds we have funding for, we're gonna have to pull from other EDs, and not only just rely on our own. Because as he had shown, as you had shown on the slide, that there's 60% of the patients probably are not gonna be going home because they have needs that are gonna exceed what we could do at the home. And, so, yeah, so that scalability is also, if you're considering doing a virtual observation unit, it is going to be highly, highly subsidized if you try to do it within one system, but you gotta start somewhere. And then that trying to scale, to scale to other EDs. Similar to tele-triage, too, right? Tele-triage has shown benefit, especially more return on investment, per se, when you scale to more than just one ED. And with the virtual aspect of it, you can, right? There's a lot of other red tape, but it's something that's feasible, so.

- You know, that brings up another point I meant to make, and one of the beauties of having the tele-observation shift, like when we're rounding, you're pretty busy. Like, you're focused on getting things done. So we protect that time of their shift. After that, we open that position to tele-EMS calls. And so it's very easy to be, like I can, if I'm calling a consultant, I can say, "Hey, gotta call you back." But we'll take, we've now done over 70 tele-EMS calls in critically ill patients in South Georgia. But it's not a consistent enough volume for us to have a dedicated tele-EMS position. It's disruptive in the ED, where you're running codes and taking care of people. It's a perfect fit for tele-observation because that position, after rounds, is pretty much available and take calls. So it's really helped us complement staffing. We probably won't be able to do, run our nurse call line because that line is becoming fairly busy. They're seeing like three to four people an hour. So that'll be a separate tele-emergency medicine service.

- Mike, do you wanna see what Kevin Curtis had said? The idea of, I think it would be more of doing virtual observation, but extending it to I think what you were calling sort of that Wild West, the type four, where these are patients that you have someone who's in the bed that's an observation patient on the window side, and the door side of the room is actually an inpatient. So had you ever, do you know of any places doing that or have you considered that?

- Yeah, you know, it does beg the question is what's the secret sauce of the unit? I really, I've kicked this question around a million times and I think the secret sauce, believe it or not, is emergency medicine. That we are very focused on time. We're very clock-focused. Having a dedicated unit, having protocols in emergency medicine now. The question is, do you need a dedicated unit? We thought it had to be next to the ED until we did a small study years ago when we moved up to the 11th floor for a month, and we found no difference in length of stay when we were 11 floors away. So you can be remote in the hospital. The next question is, could you be scattered around the hospital? I'm concerned about that because I know the med-surg floor nurses tend to fall into that once a day as opposed to once-an-hour mindset. I haven't gone there 'cause I suspect it might not work. That's my personal opinion. Now, the question, as I understand the question that's asked, is what about ED boarders, which I would take to be somebody who's admitted as an inpatient for another service. Those patients really, I mean, in our shop, if they're inpatient, they're managed by the admitting service or ICU, believe it or not, in which case we're stuck with them until there's a bed. We've started piloting having the intensivists come to the ED, manage 'em. At some of our hospitals, on nights, there's an EICU program where they remotely manage ICU boarders in the ED. So not emergency medicine, but critical care medicine at a couple of our hospitals is doing EICU overnight. The thing I wonder is why like the admitting services haven't picked up on this? And I'm guessing it's because of the length of stay. It works perfectly for an obs unit. There's positive literature on hospitalists managing patients or rounding a patient. There was one paper, I seem to recall, where hospitalists rounded in a small rural hospital, probably the size of our obs unit, and they showed that virtual rounding with an APP worked, so. I would, if you pull boarders in, rather than having, I mean, you could have emergency medicine manage it. I would advocate having the admitting service manage it, but that probably gets into local practice too much, so.

- Like, I'll throw one other thing in here, too, just in terms of, with the virtual observation unit and sort of this idea of hospital, the acute hospital care at home. Something that I'm sort of fearing about the program that we have, and that could happen in other places, too, is that our system has a very bold plan to have 222 or 224 beds by the end of this next year in the homes, for inpatient level of care. And if they aren't meeting that, and this, I'm assuming, is happening in a lot of places, right? There's a lot of bold visions. There was a lot with the new waivers out there for inpatient level of care to be reimbursed at home. And so healthcare systems are looking at this. And so I do think it's wise for emergency medicine to think about how can we piggyback onto that? One downside of this strategy is that if they have APPs and they have physicians that are hired to staff these spaces and these beds, and there's not as many of these inpatient patients that they're finding that could be appropriate for home, or consent to be at home or whatever the different reasons are, are they gonna start taking some of these observation patients away from emergency medicine and have those count towards patients in at-home, receiving care at home? And then are they gonna say, "Well, emergency medicine's not needed? Our hospitalists, they've taken care of observation patients before. Can they do that?" And, again, I always hark back to Mike, your studies that you had shown previously for in-person emergency-medicine-based observation care and all the savings in both money and bed days. I do feel like this is gonna be a little bit of a pendulum swinging back and forth a bit for those of us trying to do the virtual observation that's emergency-medicine-run. And I think the flag in the sand of why I think emergency medicine is really key in this aspect for the virtual observation, the way we're doing it, is that we're working with EMS. So for all of us EMS colleagues out here on this call, too, is that we are very closely integrated with EMS, with how much we've worked with them in the past. And so our community paramedics are used to working with us, we're used to working with them. And that may be a little different for the hospitalist side. So it's just a warning out there for those who are considering some of this and what I think might be the sign of the near future before some of this settles out too, so.

- You know, that reminds me, like halfway through the, in the first year or so of the pandemic, the CDC published a study that showed that with the pandemic, there's a rapid decline in the number of ED visits. And then they showed the number of telemedicine visits, and there was a corresponding rise in telemedicine visits. And what that showed me, which I believe, is that a lot of things that were going to the ED were being managed by telemedicine. Now, there may have been some correction along the way, but, so with that, you know, and the the workforce issues that we face, my perception is there's a train coming down the track. You can get on the train or you can get run over by it. And for my colleagues, for the people that we're training in residency, I wanna protect those jobs, those positions. I want us to be the ones on the front line, providing time-sensitive care to acute conditions that we're the best qualified to do. So. Totally agree, credentials for emergency medical control is an E/M issue. Yeah, and, you know, I get a lot of the concerns that the people don't wanna go outside of the ED, but, you know, my line in the sand is if it's time-sensitive, somebody calls and they wanna be seen, you know, the prudent layperson definition of emergency, they wanna be seen within, you know, a day or two, and they can't be seen in the clinic, we'll take that. If it's 911 critical care, we got that. If it's observation medicine, that is a time, it's an outpatient service, not inpatient, and it's defined by time, like emergency medicine, we got that. So telemedicine is like ultrasound. It's a tool that's been added to our armamentarium that expands our scope of practice or refines our scope of practice.

- I don't wanna derail the conversation, but you had mentioned the nurse triage and a separate tele-emergency service taking those nurse triage calls or supplementing them. It's like actually, my like area of research within the VA world, but this is the first that I've heard of it being done outside the VA. So if we have time, I'd love to hear like a quick two minutes about that.

- Yes, and before I answer, I wanna put in a shameless plug. This Friday, the Michigan College of Emergency Physicians is putting on their annual observation medicine conference. Yours truly will be the keynote speaker on the topic that you just heard from. But if you wanna learn more about observation medicine, it is a great crash course in one day. So just Google MCEP observation conference. And if you can't find it, email me and I'll let you know. Back to the nurse call line. So at Emory Healthcare, we have a nurse call line where people call to schedule appointments, with medical concerns. They get thousands of calls per day. It's a staggering number of calls that they get. Some of the calls are triaged, like you need to go to the emergency department. Some of the calls end up in appointment. But there's a well-defined subset where we've worked with our nurse call line nurses for telehealth-eligible conditions that we've said, "Let's transfer these to the nurse call line." We started small. We're about three months into it. We work 2 to 6, Monday through Friday. And, you know, we started with three conditions so everybody knew exactly. And I did extensive training of the nurses and the docs, how to do it. We're using a platform, I'm not, called Ola. I'm just pointing out there. It's how we started, with Epic as backup. But, you know, we started, and we're getting one or three, we're up to about three or four, and I think four's the max number of people that we can see per day. And a lotta COVID right now. A lotta people with, you know, low-risk COVID that don't need to go to the ED, that may or may not qualify for Paxlovid. No lab, no X-ray, but a lot of diverted or avoided ED visits. So it's been working well. And anybody else doing anything similar? Sata, I see you on the call. I know that you've been looking at our data. Anything that we can glean from that?

- [Sata] Yeah. Yes, yes. Our data is showing that patients are very, very much satisfied with this. And we have seen over 1,000 patients in a very short period of time through diverting, to them actually being seen fully, the full ED visit, a virtual visit instead of having to go to the ED or seek care that way. So it holds a lot of promise and we're just... You know, the interesting thing about all of this, too, is the flexibility it allows for physicians. You know, Mike Ross is talking about one physician rounding at several hospitals from the comfort of their home. And also having that scalability of what they can do, and that flexibility of then having them being able to plug into EMS calls or into tele-triage. If the ED is backed up, they can start preemptively facilitating care of patients in the waiting room. So there's a lot of flexibility, which the physician experience has been overwhelmingly positive towards. And we're seeing that with the nurse call line, too. It's a four-hour shift. I mean, physicians don't typically, ER physicians don't typically work four hours. And prior to even initiating it, we had surveys, physician surveys, and it was very interesting, when we asked about how long folks wanted their shifts to be, and the overwhelming majority said four to six. And this is something that's doable when you're approaching it from a telemedicine perspective, and it's a service to our patients.

- Yeah, it's extra duty. It's not part of the core schedule. They sign up. It's literally every time we put the schedule out, it's filled within an hour. I mean, it's like staggering. It's almost like you can hear. It's like, it blows me away. So, see. Yeah, happy to talk to you more about it. And I can definitely see the potential relevance for the VA system.

- Yeah, and I'm actually like clinically at U Dub now, I just, I do some research with the VA, but I've been trying to push our health system in that direction because I've seen how well the VA model works. So it's great to, I will certainly email you.

- [Sata] Kathy, are you the one who authored? There's an abstract that I included in our newsletter. I'm the newsletter editor for the section, and there's an abstract that I included that dealt with a nurse call line, like a nurse triage at the VA. Was that your article then?

- Quite possibly.

- Okay, okay.

- I don't know if there's been multiple recently. Well, you see, I found that out without ever having met you or talked to you, so there you go.

- Kathy, you're getting roped into some type of newsletter article, trust me, Sata will . And it was, I believe it was. I think I remember seeing that in the newsletter. I think it was yours, Kathy, so.

- Congrats, and here's your volunteering.

- Sata taught me everything I know, so she's a good person to work with. Ellie asked about staffing, do we use a subset of ED docs who just do these shifts? Quick answer is no. Over the years, we've dabbled with that. I'd rather have everybody that's sending patients to the obs unit also be taking care of patients in the obs unit. That being said, there are some people that prefer those shifts, and they tend to gobble them up. What we've done. our schedule is brilliant. Matt Keadey. We were having, every month, we have like a huge group, we were having holes in the schedule that we have to fill with the on-call position. So when the schedule came out, he decided to make the weekday, Monday through Friday, telehealth shifts, take them out of the scheduling matrix. And after the schedule goes out, make those the open shifts and then offer them for extra duty. And guess what? They get gobbled up like that because it's Monday through Friday. It's daytime. It's a nice easy, shift. It's a great way to balance the gaps in the schedule. It's a lot easier to ask people if they wanna volunteer to work at telehealth shifts than to tell them they have to work Saturday at midnight. So they filled all the unpleasant shifts and they let these. If there's ever a deficit in shift coverage, they've leveraged these as the open shifts, and it's worked wonderfully. But I hope that answered your question, Ellie. You have an APP onsite, 24 hours, to cover the unit. Each facility work in the rounds. So this gets into, and here, before we get in the weeds, which we don't have time to, there's two models. One is a single service, one is a two-service model for billing, and what they do and what they can and can't do in the ED. Right now, we've defaulted back to the single service and we're moving back towards the two service. It's a longer story than we have time for. But the APP, if they work under this model, we found that they're most busy like 7 a.m. 'til, you know, 8, 9, 10, 11 at night. So we'll have an APP, depending on the size of the unit. We staff about an hour per room with APP time. So you scale that up and down according to the number of beds. And then after that, an APP in the ED is first responder for minor issues, and the ED attending's always available for any major clinical issues. So that's how we staff. I hope it makes sense, J. Michael.

- Other questions from the group?

- Okay.

- Yeah. Well, thank you, Mike. This was, as people have said in the chat, super informative, very helpful. And I'll be really curious to see sort of what the trajectory is with sort of these virtual observation unit, tele-observation units over the next few years, especially with, at least for ours, knowing that the billing is at least available for observation care using telehealth through 2024. So those who are doing stuff like this or planning to do this, get your data out there so that CMS is nudged to make this more permanent, so.

- I feel like we've discovered penicillin and I wanna share it with the greater medical community so you can start using it to treat your patients.

- Awesome. Great. In terms of anything else, I just wanted to, I had received this from, actually our ACEP president emailed me yesterday with this link to this article that had come out in "Slate" magazine that basically is not great for telehealth. So I wanted to end the meeting, not on a sore note, but just to keep an eye out that there's gonna be always myths that we're trying to bust. So let me just throw this in the chat. I just gotta find it again. I had it copied and pasted, but then I didn't. But basically, this was an article looking at, I don't know if people have seen this yet, but they basically are saying that telehealth is undermining rural hospitals and is a reason for the closure of rural hospitals. And that they were looking at things with the Interstate Licensure Compact showing that if a state joined the Interstate Licensure Compact, that patients in that state more likely will be cared for by a provider from out of state than someone in their own state. And then they said this is closing rural hospitals. Rural hospitals were closing before this. This is my thought, telehealth is a tool. Could be used for lots of different reasons and lots of different unintended consequences. And there are programs out there that are there to try to keep these rural hospitals open and running because they sometimes are the largest employer in their towns. And so just, it's just one of many, but one that came across from the ACEP president to me, that it was like, hey, FYI. So this is what people are seeing out there. And yet another reason that, or not, another sort of thing we'll have to sort of myth bust in all of our discussions with stakeholders. So unless you're the author of the study, I'd like to talk to you separately. If you are, then let's talk. But I think when someone else was, Sata, you were looking at, or someone's looking at, it was a people in finance, there's someone from finance and then also IT. So I don't think there were any clinicians on this one, so you're safe. But, yeah, just to keep everybody, just sort of, just a little thing to, after hearing about stuff that we can really do, also know that there's still stuff we need to sort of work, unfortunately, upstream against. So on that happy or not so happy note, Scientific Assembly is coming up. Hopefully we'll see many of you in Philly. I, unfortunately, am, I'm 13 days status post-surgery, so I'm still standing up, which is great, but I am not allowed to travel by that time, so I will not be there in person. However, I will be there virtually. And if we're doing telehealth and I can't do this virtually, then we have issues. So I will be there virtually. And I really hope a lot of you come. I'm really trying to make sure that we have time built in where there's a lot of chance for mixing and mingling for networking 'cause there's so many people doing so many great things on this call, and so many people that are dealing with challenges that are similar and other great sort of lessons learned. So please plan to come. I believe it's at Wednesday, 2:30 to 4 p.m. And we'll send more information out, too. And then upcoming in November, we have Kelly Rhone from Avel eCare, formerly Avera, coming to speak on sort of the new stuff that they've been doing and what they've been doing in the emergency medicine space. So looking forward to seeing you all there then, too.