Michael Bouton, MD, Chief Medical Information Officer, New York City Health and Hospitals in conversation with Paddy Padmanabhan, CEO of Damo Consulting on the Big Unlock Podcast – “The equilibrium between in-person and video visits will be determined by specialty-specific care”

PP: Hello again, and welcome back to our podcast. This is Paddy and it is my great privilege and honor to welcome Michael Bouton, Chief Medical Information Officer at the New York City Health + Hospitals, New York City H+H, as it’s called. Michael, thank you for setting aside the time. And welcome to the show.

MB: Thank you very much, Paddy.

PP: Would you tell us a little bit about your organization and your role within the organization?

MB: New York City Health and Hospitals or H + H for short is the nation’s largest public health system. We have 11 acute care facilities, a long-term care facility, a couple of SNFs. We also have over 60 clinics in all five boroughs across the city. And I am the enterprise Chief Medical Information Officer and also a practicing emergency medicine doc at Harlem Hospital, which is one of our hospitals in the system.

PP: Thank you for that background. So, Mike, New York has been one of the hardest hit by the pandemic, and we’ve all seen the numbers and everything. Tell us a little bit about some of the changes that your organization has gone through in responding to the pandemic over the last few months.

MB: The pandemic did a bunch of things, one of which is that it accelerated plans, it accelerated some of our digital transformations. It brought us to a place where I thought we might be two years from now, but we were there in three or four months during the pandemic. A couple of good examples are, in our ICUs and our ED’s, we have vital sign monitors that are linked to Epic, so vital sign monitors can be on patients and it transmits to a central station. The nurse at the central station can monitor the patient and then that information automatically kicks into our electronic medical records. So, it’s a great time saver for the nursing staff and it also allows them to monitor up to 36 patients sitting at a station. And in our EDs and ICUs, we’ve had that since we went live with our new electronic medical records in the past few years. However, on our medical floors, our surgical floors, a lot of other units in our inpatient side of our hospital, we did not have that ability. One thing that became clear very early on, in the pandemic, was that the COVID patients were going to require continuous pulse oximetry. We were sending everybody home that had a pulse ox, 95 percent or greater, which I think is appropriate clinical practice. That means if you were getting admitted to our hospital, basically you are hypoxic, and the patient would sometimes deteriorate quickly. So, they really required continuous pulse oximetry. We did a whole bunch of things, telemetry devices cannot do just the rhythm strip, but they can also do pulse ox. And then we went out and bought hundreds upon hundreds of vital sign machines that could do this continuous pulse oximetry. Our medical floors were transformed in the matter of about six weeks from places where our nursing staff would go bed to bed, taking vitals to not having to do that and that was part of our long term plan. That’s something that I had wanted to do for the past year. But from funding, time, and effort, and where we were going to prioritize that shot up very high on our list. Another example is our telehealth, our ambulatory video visits. That’s one thing that we had started with, in a non-integrated fashion, meaning we would basically send a link to a patient, and they could click on it and come in and it was very sparingly used. That whole process has gotten accelerated. For us, the tide of the pandemic was in April. We were doing almost all of our visits either via telephone or the video. That was the transformation we never fully expected to go almost all of our visits. We didn’t expect in-person part of our ambulatory side to go away entirely and it did. It dropped to almost nothing for a couple months.

PP: I want to spend a minute more on the vital signs, the telemetry, the Pulse ox problem that you had to work quickly. I imagine that if you had to go out and buy hundreds of machines integrate them, reorganize your processes, train everybody, and so on. That must have been a gigantic lift for an organization as big as yours. How do you manage it?

MB: It was a gigantic lift. We developed a dedicated team and prioritized it. What we had been working on previously was integrating some of our procedural areas. So, our GI suites were all fully integrated. We took the teams that were doing that, and we focused them on our medical floor. We took anybody who was able to do this integration work and we made this our enterprise’s number one priority. We went from acute care center to acute care center, outfitting the site CMIO, CMO, CNO, like the site leadership. So, from one of our 11 acute care would tell us which unit they wanted integrated first and we would come in about a day. We would get that whole unit integrated. We would do it off the floor without exposing the IT staff to the COVID patients on the floor. The devices would be brought up. We would stay there and troubleshoot, and move on to the next facility the next day. Now, training the staff was difficult because training in the best of times takes time, and doing so in the middle of a pandemic when everybody is already stressed. Everybody already feels like they have too much work because they did. We had a lot of COVID patients, and it was stressful. So, trying to train people in that environment was hard. The thing that we had going in our favor was that this really was a time saver for the staff. So, people that saw the benefit invested the time upfront. But getting them to pay attention for that first couple of minutes, that was challenging. And frankly, that is why we’re continuing the training at this point. No one knows if New York’s going to get a second wave, but we certainly are preparing as if we are going to. Not every single bed in our hospital is capable of continuous monitoring at this point. But we continue to expand our number of beds that can do this. And really, I think what’s more important than adding those devices is that training these and getting our nursing staff and our PCAs, which are our patient care, associate with them. Getting them able to do this is what we are working on right now. It’s really where the rubber meets the road.

PP: I heard someone say that, 80 percent of these kinds of programs, even if they’re technology programs, is really about the people. 15 percent is about process and the remaining 5 percent is tech. And the tech is the easy part. Everything else that surrounds it and putting it in place and making it work seamlessly, that’s where the lift is. And that’s kind of what’s coming through from your comments as well. Let’s talk about telehealth. In the wake of the pandemic, telehealth was forced upon us. Prior to the pandemic we had all been making progress. Every healthcare institution in the country was adopting telehealth to some degree and some were a little further ahead than the others, but there was some progress. At a broad level in the first few weeks of the pandemic practically every institution that had telehealth capability kind of blew through the previous years’ total numbers within the first couple of weeks in terms of just the total visits that they had to manage, the telehealth platforms. Now that was all very emergency driven because of the pandemic. But three months, four months in what are some of the challenges that we are now having to address, having had to put in and accelerate telehealth adoption. What are some of the challenges that you’ve had to address and can you talk about one or two learnings?

MB: Sure. This is also relevant to what you were saying for the vital sign monitors. This is not my key project, telehealth is not my key project. It’s a clinical transformation. So, this is not just rolling up the technology and say, hey, you can do a video visit with a patient now. It’s all of the little workflow things that you didn’t. A couple of things that I’ve seen be successful with others that we are implementing now, they sound small, but I think they’re really important to the user experience. When I say user, I mean the patient and the provider experience. It’s virtual ruling. I remember when I started working at New York City Health and Hospitals, a decade ago, sometimes I would have to go out to the waiting room to call my next patient. We realized very early on that was entirely inefficient and we took that responsibility away from the doctor so we could work at the top of our license. But when we rolled out the video visits, we basically did the exact same thing. We had the doctor initiating the call with the patient, makes more sense in my opinion. It’s having really anybody else, if you have a medical assistant or registration staff, you can have them initiate the call with the patient. And if they’re having trouble, or if the patients having trouble getting on, it should be the MA that calls the patient, not necessarily the doctor. This isn’t different than if I was the one going up, had to call my patient from the waiting room and they weren’t there having to wait three times and go back a couple of days to go back. It would be a waste of my time. But if you can have somebody else do that and you can have your provider seeing another patient or finishing charting on somebody else, it is just a better use of resources. The thing is, we have those resources because the in-person volume went down. So, we had excess capacity of those ancillary staff. It’s really about leveraging mode. And another ancillary staff is that it’s really critical for New York City Health and Hospitals, those are important to a lot of organizations, is interpreters. How do you get an interpreter on the video visit? It’s something like 30 percent of our patient population is not English speaking primarily. So, that was absolutely needed. So, leveraging our in-house interpreters. It’s not just the video interpreters that we can call online, but there were problems integrating with those and we’re working diligently at that. But about using the resources you already have on site to help you with the technology. And that’s where we found the most success in this medium-term period.

PP: Now, related question. And this question probably is relevant in the context of the vital signs monitors as well. Ultimately, you have to integrate the data that is coming of your backend systems or of the devices that you put in front of patients or caregivers as the case maybe that’s kind of your world, I imagine, in many ways, CMIO. So, help us understand the data integration challenges. We know that there are several and they’ve been around since before the pandemic. What new challenges did it create for you and what new opportunities did it create as you really transformed the way you deliver care and interact with patients? Can you unpack that a little bit?

MB: Sure. The challenges when you’re talking about the data flowing in through the monitors: if the nurse is going to bed to bed and they write down the vital signs, come back over to the system and they type it. There might be a manual input of data, meaning that they might actually just type it in incorrectly. But otherwise that data is pretty accurate. If all they are being asked to do is verify that the information coming from the monitor is correct, they just being human like everybody. They might just say, yep, that looks right. And they’ll know the pulse ox might be reading zero because, 70 something percent, it’s not attached, if the patient got up to go to the bathroom. So, you do have various data entering the system, which is problematic. I got to tell you; I did not see as much of that as I thought. I think our nursing staff, there was a heightened level of awareness for our sick patients and they are the ones that were getting these continuous vital sign monitors. So, while certainly it’s something to watch out for, didn’t turn out to be a major problem for us. The opportunity, which is the other side of this, is that we put in a machine learning algorithm that took a whole host of factors. So, a patient’s age, patient’s diagnosis, their vital signs that were coming directly from these machines, their lab values. It actually was able to predict if the patient was going to deteriorate, meaning get intubated, be transferred to the ICU, or die in the next 24 hours. And using this as we use other early warning system to bring more critical attention to the bedside. And that was a major win for us, our staff liked that. In the face of an increased patient volume, we needed that more than ever.

PP: So, this is a great example from a clinical standpoint. So integrating the data to trying to do in real time, you’re running a machine learning algorithm on it that can provide you with some predictive values that help you target the patients most at risk for deterioration, and intervene in a timely manner and save lives. Great example. What about the telehealth side of it? How did you integrate the data? There is an administrative side of it. For instance, you mentioned that you used to send out a link. People would get on and do a video call. How would you link it back to your billing system as an example? Make sure that you capture encounters and the billing put it appropriately. And then doing it all in a HIPAA compliant way so that privacy and everything is taken care of. You talk a little bit about the access side of it.

MB: I think we’re speaking specifically about video visit here. But a related issue would be the incorporation of digital and digital vital signs, digital information coming from the patient that’s not a video visit. So, I’ll hit on that in a second. But in relation to the video, we had patients log in through their portal, primarily. I mean there are a couple of different ways that we did this. During the pandemic, I know somebody who likes to roll out an enterprise solution, get it adopted widely, and have really a single way of doing business. That wasn’t really a possibility during the pandemic. We had to use a bunch of different forms of technology to meet our various use cases. Speaking specifically about ambulatory scheduled visits, we use our patient portal, a large percentage of our patients had already an app on their phone. And if they didn’t have that app on your phone, we required that to put it on to have a video visit with us. Now, people that weren’t able to do that, we provided another means of coming in. But that was our standard way of doing business. And then from a billing perspective, what we realized, and what is not fully my area of expertise, but we did telephone visits and we did video visits. My understanding is that video visits were reimbursing about three times the rate of telephone. So, there was an economic push to move towards video. I think there is a clinical advantage of a video visit. The question is, how much advantage is that? I clearly want to be able to see my patients, then just eyeballing your patient, it has a lot of value. You get a sense of their respiratory rate. But we have both clinical and financial push to move towards video visits.

PP: What about other structures, your in-patient, virtual care, tele-ICU, and stuff like that. Maybe you could touch on that.

MB: We were a little bit further ahead with that for the pandemic, we had more experience. I think its easier to scale something that you’ve already had worked with, than rolling out something entirely new. And that’s true with every project I’ve ever done. This was really no different. So, yeah, we did a tele-ICU. We made it easier for folks to log into the electronic medical record from home and actually see the views of their patients that would be most beneficial to them and specifically in this case, to the intensivist, and then gave the ability for a video interaction. Now, I think video interactions are very helpful with ICU and we clearly did that. But interestingly, the video component in the ICU, specifically when I talk to my intensivists, it was valuable, but not nearly as valuable as I thought because so much of that data was already in the system. And ICU patients have such rich amount of data in the system, you have your event settings, your vital signs, your lab, your nursing. You have so much available in the system already that you don’t need to look at the event if that data is already in your electronic medical records.

PP: Interesting. So, back to telehealth and I want to touch on one more thing that is remote monitoring. Again, this is part of your world. You are taking care of your patients who are out there with chronic conditions, who are not necessarily coming into the hospital. You’re tracking them through devices and wearables and so on. Can you tell us a little bit about how any of that changed? And where do you see that heading in the wake of the pandemic?

MB: Yes, this is such a rich area to move into. I think there’s a huge amount of benefit here. I think most of your listeners will be enthusiastic for some of these specific projects that we worked on. If you have an implantable defibrillator, and you’re at home, you have an AICD, you don’t necessarily want to bring those patients into your hospital to get that device interrogated. Sure, if you’re sick and you need to come in during the pandemic, we want to take care. But if we could do that remotely and if we could get that information from your device without you having to physically come in, that’s a clear cut win. So, we’ve got a lot of projects like that. Now, the other very clear use cases for this are the tracking of your diabetic patients, tracking your hypertension patients. And to me, I think we never want to get rid of the in-person experience altogether. I think there’s a real value in doctor-patient relationship. Face-to-face interaction between hands on the patient, even if it’s not the most clinically beneficial, it has a therapeutic advantage. We’re not looking to get rid of that. But if I’m seeing one of my diabetic patients four times a year and now all of a sudden, I can look at their glucometer on a weekly basis and see how they’re doing and have machine learning algorithm seeing in the background, notifying me when things start to not look so great, that’s a clear cut win. And we’re there. This is not something that, we need more technology for. We are now at the point where we can do that. It’s about developing those workflows. And then what are you going to add to flag a patient like great. This patient is at high risk. But then what do you do? Is a phone call enough? Do you need to schedule for a diabetic, you need to schedule them with a nutritionist? Can you do look at their labs? And I think you it’s going to need to come in and get more lab work. We’re yet to see a whole lot of really rigorous studies on when X happens. This is the intervention that you should do that is clinically proven. And I think that’s really fertile ground for research.

PP: Yeah. Well the hypothesis here is already being validated through marketplace activity. We just saw the emergence of two big companies in this space, Livongo and Teladoc. One is primarily on the virtual consult space. And the other one is in the remote monitoring space and creating a gigantic entity, which kind of is a validation for the opportunity in this area. At the same time, I want to kind of explore the contrarian view a little bit as well. The recent data seems to indicate that telehealth visits have dropped off a little, maybe because they swung too much to one side in the wake of the pandemic. And to your point earlier, nothing replaces in-person care and for certain kinds of needs. I’d call it as an example, it’s hard to manage a condition entirely, remotely. So, we are seeing some of the swing back. But there are other issues related to access for rural or indigent population that may not have the broadband access, that don’t have devices, smartphones and things where you can jump on into a video console. Where do you think we are headed in terms of an equilibrium? Maybe from the point of view of your world at NYC H + H where do you see the equilibrium and what could be the roadblocks you have to overcome in order to really realize the full potential of telehealth and remote care models?

MB: I think it will be specialty-specific, meaning how much is in-person and how much is remote. If I’m going in to see my orthopedist because my knee hurts, there could be a lot of manipulation of my knee. And I think they are going to really need to feel and look at it. I think you’d be able to maybe lessen that on a video then, not none, but less. If you’re going to be seeing your primary care doctor to manage your hypertension and you’re coming in every four months because you’re having trouble with that. Well some of those visits can be done remotely and it can be done remotely without losing a whole lot. I think that the annual in-person physical is also going to be generational to certain extent. When I have the ability now to do video visits, and I can instead of taking a half a day off of work, I can hop out for 20 minutes on a call and then go right back to work. That’s what I personally would want to do as a patient for the foreseeable future. But if I need to go in and get bloodwork done anyway, well, I’d rather just do the visit in person. I’d like to see my doctor. So, it’s a question of the opportunity cost and what the patient is giving up. For example, if you have to come in any way to your ENT to get a scope done and you have an ENT who is going to be doing a scope on me. No, of course, I’d rather come in and get my blood work done and see my provider, all at the same time. But for those remote visits that are amenable to a remote interaction that don’t require physical contact, I think a lot of those are going to go away.

And when they’re settled out Paddy, I have no idea. If we switched to 90 percent video during the pandemic, let’s just say we were 100 percent in-person before. My guess here is we’re going to see maybe 20-25 percent of our visits on a video basis when the world goes back to normal. But after we get a vaccine and people feel pretty safe going back to their normal life, we’re going to see a tremendous increase in our video visits as compared to our baseline of six months ago. I don’t yet see it being the dominant trend. I think what we will see is more opportunities for low-intensity care interactions, which are – “Hey, what happened with your blood glucose? I saw it hit 400 today. What happened?” I could see that the social worker reaching out, but not replacing the doctors. I think it would increase the quality of care.

PP: I think whatever the next normal, it is going to look like what I kind of agree with you. I think we are going to see the needle shift towards virtual care models for many types of care. But then it’s especially dependent. I would imagine that more of ambulatory care and more chronic care is going to be amenable to virtual care models than procedures and things like that. But even for certain kinds of care, to your point, to see an orthopedist talk for oncology as examples. I want to leave you with one more question, which is, as they say – ‘never waste a crisis’. COVID-19 is definitely a crisis, but it’s also an opportunity. Where do you see the biggest opportunity in your world as a CMIO of New York City H + H, where you see the biggest opportunity arising from this crisis?

MB: So we are pretty large system, implementation of our new electronic medical records spanned a few years and we finished in March. So, we put in our long-term care facility in March, right before the pandemic started. It was great timing. What we used this crisis for was to act as a system, meaning we had system level data that we just didn’t have before. When we compared the capacity to bed capacity at one hospital towards another. We were really comparing apples to apples. This allowed us to transfer patients from the hardest hit hospitals to the less hard-hit hospitals and really have a fair basis in comparison for why we were doing so. This improved patient care. I think the strength is to save lives, because if you got a hospital that was way over capacity, we could get them out somewhere else. And the other receiving hospital had all of their data from our other hospital, which is a huge benefit to our system and eased the transfer process. But what I’m saying about acting as a system, the literature of the COVID pandemic changed at a dizzying pace. So, was hydroxychloroquine good or was it bad? There was a time where people thought it could actually do something positively. And I think that’s changed. But there are things that have been proven very effective. So, like dexamethasone, we have a randomized trial showing that it works. So, we have in our order sets, as soon as those trials came out, we added in dexamethasone.

MB: And, there’s a million different examples just like this. But when we made those changes, we weren’t making those changes at one of our hospitals. We made those changes at every single hospital. The reason it was so easy to do for us is that we only have one order set for this. So it drove us towards an enterprise standard and enterprise way of doing business. When we put into vital sign monitors, we didn’t all eleven of our hospitals weren’t buying their own vital sign monitors. We were buying them for them. So, we already had a clear path to integration. So, on the others, these examples and so many more. But we got to act like a system really came together. We were able to achieve more. I think that would best take way for our organization. That standardization allowed us to do more and deliver a better product to our patients and to our hospitals better.

PP: Fantastic. I think that is a fantastic take away. I’m afraid we have to leave it there, but such a pleasure having you on and look forward to following all your work on NYC H + H and all the very best to you and your team. Thank you for being with us.

We hope you enjoyed this podcast. Subscribe to our podcast series at  www.thebigunlock.com  and write to us at  info@thebigunlock.com

Disclaimer: This Q&A has been derived from the podcast transcript and has been edited for readability and clarity