In this interview, Anthony Guerra, Founder and Editor-in-Chief of healthsystemCIO, sits down with Jeremy Meller, CIO of Children’s Healthcare of Atlanta (CHOA). Jeremy provides insight into his organization’s recent achievement – the construction of the Arthur M. Blank Pediatric Hospital, a state-of-the-art facility that spans two million square feet. He highlights the unique design considerations – like private rooms and ample space for families – and the innovative technologies that were integrated to enhance patient care and staff efficiency. From real-time location systems (RTLS) for better staff security and workflow optimization, to the deployment of an unprecedented fleet of 90 robots dedicated to logistical support, CHOA has embraced a forward-thinking approach to meet the increasing complexity of patient needs.
LISTEN HERE USING THE PLAYER BELOW OR SUBSCRIBE THROUGH YOUR FAVORITE PODCASTING SERVICE.
Podcast: Play in new window | Download (Duration: 31:55 — 21.9MB)
Subscribe: Apple Podcasts | Spotify | Android | Pandora | iHeartRadio | Podchaser | Podcast Index | Email | TuneIn | RSS
Anthony: Welcome to healthsystemCIOs interview with Jeremy Meller, CIO at Children’s Healthcare of Atlanta. I’m Anthony Guerra, Founder and Editor-in-Chief. Jeremy, thanks for joining me.
Jeremy: You bet, Anthony. Thanks so much for the invite.
Anthony: Great. Do you want to start off by telling me a little bit about your organization and your role.
Jeremy: I am with Children’s Healthcare of Atlanta. I’ve been with Children’s for about 12 years. I’ve been the CIO since 2000. Children’s Healthcare of Atlanta is one of the busiest freestanding pediatric organizations in the country. We have three hospitals, around 800 beds, about 20 locations, and about 15,000 employees. And we’ve recently just opened the Arthur M. Blank Pediatric Hospital, which is a 19-story, 2-million square foot, $2 billion new facility here in Atlanta, and we’re really excited about it.
Anthony: Wow. Excellent. About how long has that been in the works?
Jeremy: The concept is probably about 10 years old. We had attractive property with about 16 one-story buildings on it, which was our administrative campus. We then spent the next several years purchasing land and businesses all around that to expand that footprint. We then built a two-building administrative center, and a conference center, which we call the support center. Then we were able to move everybody out of those buildings. We broke ground in early 2000, right when the pandemic started, and started building our new hospital.
Anthony: So early 2020, is it?
Jeremy: Early 2020. About four and a half years of actual construction. We had about 10,000 people involved in the construction, as you might imagine, had to do it. Started right during the beginning of the pandemic so it was challenging, but we opened it about six months early in comparison to the original schedule.
Anthony: Very exciting. I think you said 9/29, I have down here when it opened, September 29th. So very recent. And I believe there’s a number of technologies that you’re utilizing there. If you want to start off wherever you’d like and tell us maybe some of the things that you’re featuring there that you put in place.
Jeremy: One of the first things we did was we said, okay, what are some of the biggest challenges that we’re going to have? And one of the things with modern facility design, and you’ll see this, is just about all new hospitals that open these days have, even in their intensive care units, have private rooms. And private rooms that are a little bit more expansive, more rooms for family and visitors, et cetera. And what that ends up doing is really blowing out the size of your facility. So that’s how we got to a 2-million square foot facility. Our hospital is the same size as Mercedes-Benz stadium here in town, from a square footage perspective. So that’s great from a family experience perspective, but it’s not great in terms of time and efficiency of moving around the hospital.
One of the first things we learned was that the facility size was going to be an issue and we were going to have to find ways to save steps, both physical steps and process steps, in order to help our clinicians maintain their efficiency. There’s a group of technologies that we implemented behind that. Then the next was the fact that we’re going to have and continue to have among the most complex patients in the region and in the entire country. With the amount of patient complexity comes just overwhelming amounts of information. And so we said, okay, we have to do something about how information is managed. How can AI be used, for example, to help our clinicians make better informed decisions.
And then certainly, there’s been no stopping the increase in consumer expectations, the ability to have self-service as much as possible, to review provider information in real time, and all of that coming together in the patient experience was the third pillar of where we wanted to focus in order to make a meaningful difference. All the technologies that we implemented were really, for the most part, focused around those three categories.
Anthony: Very good. You started with the pillars, and then you went from there to the technologies that fall underneath. I have a list of some that you’re utilizing, and if we can take them one at a time and talk about which bucket they go in. We have robots, location awareness, AI, patient engagement, video and mobility.
Let’s delve into a little bit about that size issue. A huge facility. You’re saying getting around could take a lot of time. What can we do to save people from having to move around the facility too much? And you probably have robots and location awareness that fall under that. But that’s an interesting approach you took.
Jeremy: What you have to do is map some foundational services to where they could lead in terms of capabilities. For example, having a really good wireless network that can handle a lot of load and performance, and having nurses and clinicians equipped with mobile smart devices, those two things together enable a lot of mobility. What we did next was really deep device integration. We have about 1,500 monitors and medical devices in our environment that are deeply integrated now. They hadn’t been in the past. And historically, you have IT and OT. OT being those things that are a little bit more either mechanical, working a little bit more in the physical world, but often they have, they’re also IT is what I would argue in many cases. And that’s how a lot of medical equipment is, whether it be scanners, MRIs, vents, or in-room patient monitors. There’s a lot of information on those that can be used in more meaningful ways than just as a device.
We’re bringing that into our network, and now our nurses, for example, on the go are able to see on their mobile device when they get an alert, and we know that better than 90 percent of nurse alerts are false alarms, they’re actually able to see the waveforms. They can see, for example, that a lead is disconnected and that can save steps. Then if they’re in the next room and they’re in the middle of a complex medication administration, they’re able to prioritize. They don’t have to create an unsafe event by stopping that action and addressing something that doesn’t necessarily need to be addressed. So that’s an example of how we’re using the technologies together.
Then you layer in things like location awareness. Location awareness is of course your RTLS system, your real-time location system with monitors and sensors throughout your organization, but then we added location awareness for staff as well. So all of the staff wear badges that have a button on them. They serve a couple of purposes. One is in the world we live in, the staff are very interested in this, their staff duress. If they push the button several times, security gets alerted to their exact location and security can respond immediately. That’s something that our staff really enjoys. We are worried about staff acceptance, where they’re going to say, ‘oh, you’re just tracking me. I don’t want to accept it,’ but we found better than 90 percent adoption of these badges among our clinicians because they want that security feature.
Then what we get with that is we can understand where the clinician is. For example, when a nurse walks into a room, there’s a halo light, there’s a board on the outside that has information about the patient. There’s what we call a halo light that lights up green, and you can look down the hall and you can see which rooms a nurse is in. And there’s a different color for a physician. You may have a physician on the unit and just with a glance, you can see where the physician may be or what room that physician may be in. That’s another example of something that can save steps. Again, whether it’s physical steps or process steps, that’s our goal, is how can we save steps, how can we use this? And that same location information is used in a number of different ways.
On the patient engagement side there’s two displays for the patient. There’s actually another screen for the parent. The parent has their own TV in the parent’s area of the room, but there is a main traditional television display, then there’s a patient information board that’s next to it that replaces your whiteboard, but it’s a smart whiteboard. And on both of those, when people enter the room, it tells the parent or the patient and the family who has entered the room.
Anthony: That’s some really cool stuff. I love the lights and the changing color. People are trying to work with these technologies in existing hospitals. I’m sure in addition to you implementing these things in your new construction, you’re probably implementing them to some degree in your existing facilities. What’s the difference there in terms of your ability to do things quickly or do things in certain ways? In what way does a new construction opportunity allow you to do things differently?
Jeremy: Great question. I’m actually going to weave in a little bit of information about the robots into this as well, because it’s very much related. We have 90 robots in our hospital. We are told by our vendor, who’s the biggest producer of hospital robots, that we have more robots than anybody in the world. We actually have nine elevators just dedicated to robots. And then on every floor around those elevators are staging areas for the robots. You go down to the basement, there’s storage and recharge areas for robots. There’s areas where the robots go through and get cleaned, or the carts get cleaned from the robots. There’s areas in food services, in pharmacy, in laboratory for the robots to sit, to be staged, to be loaded, to be sent up to the floors.
Clearly, that would be very challenging in an existing facility to build that out. Where would you get the space for that? So this was from the ground up facility that was built for robots in addition to humans. And that is already providing great benefit. We’re able to speed the time to delivery, consistency of service, et cetera.
A new facility gives you some of those capabilities. It’s not all positive. When we started planning this facility, the technology landscape changes quickly. And we thought, all right, we were going to have to accommodate, and we seriously thought this, self-driving cars and drones dropping off supplies. And these were the things that we thought we needed to plan around. And it’s just interesting because some of those things haven’t really quite come to fruition yet, but there’s other things like AI and video and video combined with AI and other types of things which have progressed faster than we had anticipated a few years ago. So there certainly have been some limitations, but some of our big bets are things that we were able to do that we otherwise wouldn’t have been able to.
Anthony: You bring up a great point. It makes me think of some devices don’t have as long a shelf life as you thought. Now, for example, an original USB charger, this piece of equipment is far outliving the technology within it because nobody uses that particular port anymore. It’s an interesting challenge from a technologist’s point of view involved in this type of construction. Where are we making bets? How do I make some sort of space where instead of having a fixed port, for example, back to my analogy, that it’s interchangeable as times change. And the equipment can continue to function, even though I think this nook and cranny is going to change pretty quickly. More thoughts around that?
Jeremy: I think you’re spot on. The other thing we really look at is this a technology that is maturing fast or not? Or is it really kind of established, in a place of maturity? And the actual video camera itself, for example, is a pretty mature technology. And now what you do with the video? What you do with the video output is pretty dynamic and there’s a lot of things you can do with it. So we were less concerned. We said, “you know what, we just want to buy a really good pan-tilt-zoom camera.” And at first, we thought, all right, and this was actually a late add. This was like a million-dollar add to the whole deal. And so that was a bit of an issue.
And then of course you have people who are like, “well, you can get these pan-tilt-zoom cameras on Amazon for about 280 bucks. That should probably solve your problem.” But then when you actually do the actual bake-off between all the manufacturers, you find that you really have to pay, but you also find that there’s been low change over the last 10 years in these cameras, the actual camera itself. Now, what you do with it, again, what you do with that video feed, you can do a lot of things with.
There were a number of cases where we said, look, we just want to get the technology in. We’re not going to have all the use cases that we know that we’re going to be able to do. And there’s also going to be limitations on the amount of change that the organization can absorb. That’s another important element I really wanted to talk about because we had to say, look, for example, we knew with the cameras, we started with two use cases. The cameras in the rooms, they turn away when they’re not in use. When you call into the room, it’ll appear on the television and the patient or parent can answer it with the pillow speaker. Now there’s an override function. If for example, there was an emergency, we just need to get in and see in the room, that’s also a functioning capability. But we said, we’re going to go live with two use cases. We went live with remote interpreters. And we have obviously a lot of patients who speak a number of different languages. And so, we wired in our interpretive services through that. That runs through there and remote family conferencing. Let’s say you have a family member, or grandma or grandpa lives across the country, you can bring them into the room with a really good, nice experience as well.
Those were the two we started with. Now everybody asks, ‘well, what about remote sitting? What about remote nursing? What about remote consults?’ Those are all things that we know that we can do, in addition to some of the AI things you can do. You could actually manage and help manage and understand patient movement, motility, et cetera, with AI now. But we knew all those things were things that we would probably do. But to get the operational leadership to really buy in that we were going to completely change the world day one, we had to agree to a little bit more of a limited scope.
Anthony: Very interesting. I’m assuming you leverage some type of governance to find out what the clinicians and the different users, where they really wanted their lives improved, or what would, for example, if you could have three things changed about that where technology might impact your workflow. And the nurse would go, “oh,” and you hear from 80 percent, this one thing bothers all of them. And you say, I’m the CIO and I know stuff and I know there’s something for that. Matching up clinician needs with your initiatives, takes a lot of governance, usually pretty formal. Can you tell me how that may have factored into the decisions you made about where to focus?
Jeremy: Sure. That was infused throughout. We had some engagements with hospital planning folks that would help us understand if the things we wanted to do was the latest and greatest. What were some of the new things that were actually happening? And you can kind of see that. There’s a number of these things which are happening increasingly. Every new facility that opens now seems to have some more of this patient engagement technology. Some of them, not all of them, have some of the robots. That’s a pretty big decision and a pretty big investment. But there’s other things that seem to be just happening. This is the best practice. For example, one thing I think about is the fact that almost all hospitals now that are new have basically what looks, a panel on the door, on the wall outside that is like a two-way box. You can open it from the outside, you can put meds or supplies or a meal in there, you can shut it, and then it can be opened from the other side so that you’re not necessarily having to go into the room and disturb the patient when they might be sleeping or at odd hours but you can still get supplies or something else into the room.
There are a number of those kind of things that we just said, ‘look, those are best practices, let’s just take those. Those are, it’s easy to put those in.’ Then there was, ‘okay, what do we want to do from the facility size? What do we think about in terms of patient complexity?’ One of the things we did was we built out what we called a cardboard city. It was a 100,000 square foot warehouse that we filled with cardboard walls that were movable. We actually modeled out all of our different types of units. Then we brought all sorts of medical equipment in there. Now we didn’t bring an MRI in, but we did make a big cardboard MRI. We brought all our monitors, our ECMO machines, our other types of machines. And what we discovered in some cases, it was everything from where the plugs were located on the walls to how the doors were positioned or that this room actually needs to be twice as big. Yeah, it technically fits everything, but you can’t also fit 12 people in here.
And it was those sorts of things that we went through that had a huge impact on the design. And then we got to, “okay, what were the process improvements that we wanted to make and what do those look like?” And we did that. We had a pretty extensive process to try to balance where we think technology is going here with auto, with self-driving cars. Is that something we have to address or not? We thought at first, we were going to have to address that and eventually we said, even if that happens, we’ll adjust to that. But there’s some other things like just the changing nature of healthcare. And over time, the acuity has, and especially in pediatrics, increased and increased and increased. And more patients that can be handled at home or don’t need to come in the office or can be handled or their diseases can be better managed don’t need to be coming to the hospital at all.
What you end up with is higher acuity in the hospital. And that’s where you get to, all right, there’s just going to be a lot more information that we’re going to have to absorb, manage, et cetera. So when we think about how the human brain works, what we want to do is offload as many of those routine tasks that take up space and allow computers and automation to do those things because they’re actually very good at doing them and reserve as much of the human capacity, the human brain capacity to the complex decision-making judgment and other types of activities that are really going to help result in a better outcome for the patient.
Anthony: That’s exactly how I think about, like, my phone and my reminders app and my calendar app. I think of it as offloading.
Jeremy: That’s right.
Anthony: It is an interesting way to put it. It sounds very accurate. As I was thinking, building a new hospital, what do they consider a shelf life for a hospital – 50 years, 100 years, 75 years? And once you start thinking like that, you say, now if I’m the technologist helping to build something that’s going to have a 50 minimum year shelf life, that’s like, we get into science fiction, like should I go watch some Star Trek? Should I…
Jeremy: Right.
Anthony: Because you get really, you can’t think that far ahead, as you said. It’s all about what construction can’t we change? You’re not going to change the size of the room. It’s not going to happen.
Jeremy: Right. Right. There are walls that are movable.
Anthony: Right.
Jeremy: And there’s walls that are structural support walls. And so that’s how you have to think about it.
Anthony: Conversations with my wife when we were buying a house, right? What can we change? You can’t change the street, can’t change the town, but you could change a lot. You could change, theoretically, you’re not going to change much of the structure. Any thoughts around that? I mean, in terms of this planning for something that needs to be functional 50 years from now.
Jeremy: It’s a great question. One of the things we did, for example, is that every room is exactly the same size at our new hospital. What we did is we said, we’re going to standardize, regardless of the type of patient, we’re going to standardize the room size. We say, okay, you know what? One of the things that does do is it makes your facility larger. We can certainly have maybe saved a little bit of space in some ways if we didn’t treat everything exactly the same. But because we did, we can very easily convert a general patient care unit to an intensive care unit or vice versa.
We also built four extra floors. We have four floors that have nothing in them at this point, but we built all the infrastructure to be able to serve all those floors. Even interesting infrastructure like our tube system. In addition to the robots, we have a very complex pneumatic tube system. And the amount of pressure you have to push through a pneumatic tube system to get a tube up 19 stories is a lot. There’s nothing new about pneumatic tubes. They’ve been around since the mid-1800s. But this is smart, automated, has location tracking and awareness, so we can keep track of where the tubes are and all of that as well. But all of these things we built so that we could expand, we could adjust, and we could modify in order to change whatever the mix of patients are that we end up having.
Anthony: Pneumatic tubes on steroids, we’ll call them.
Jeremy: Yep, pneumatic tubes on steroids.
Anthony: Any more cybersecurity thoughts? Anything different about new construction when it comes to cyber?
Jeremy: Certainly it was an opportunity to make sure that we did things right, that we had the right kind of segmentation, that we had the right kind of physical security controls. In the olden days, we had doors with keys and locks, and everything has complex security, zone security, badge access security. From a physical security perspective, that was certainly an improvement for us. The amount of medical devices that we needed to clear, we didn’t just bring things over either from – because we shut down one hospital. We moved a couple hundred patients to the new hospital when we opened it up. And so, we didn’t even roll anything into the new hospital. We had to make sure everything was cleared. We understood it. We understood the security profile because we treated it as though it was a clean environment.
We also knew that we were going to be a target. We got several phishing attacks that were aimed at our staff, that would say things like, “welcome staff to the new hospital, we need you to click here to update your information.” And so this is the intelligence, the social engineering intelligence that’s behind some of these things, but we anticipated that. So we really worked hard at training and educating our staff to stay on guard and stay alert. We are going to get cyber attacked. We ran extra penetration tests. We increased our level of monitoring with our SOC, our Security Operations Center, and we had a physical on-site penetration at test, where we hired somebody to come in before we had actually officially opened the doors, but everything was running and operational, as you might imagine, for weeks before the doors are actually opened. And they came in, they tried to plug into ports, they tried to connect through different wireless networks.
Anthony: That’s great stuff, Jeremy. We’re just about out of time. I just want to give you an opportunity for a final thought. Let’s get your best piece of advice for one of your colleagues, and let’s say they’re just at the beginning of this journey. They’ve just been brought into the meeting where they’re told that we’re going to build a new hospital, and this is going to be a long process, and you’re going to be involved. And what’s your best piece of advice navigating this journey for them?
Jeremy: You’ve got to work with your colleagues who are probably going to be really overwhelmed with the amount of change. Most CIOs are trying to move our organizations forward. And it’s not that the operational teams aren’t. It’s that they’re often just trying to get through the day and trying to get through it safely. Because if you can’t provide safe care, you can’t provide effective care, you won’t be able to do any of those things anyway. You have to live in both of these worlds and you have to work with your operational partners on what we can accomplish, what will be, with the staff. And we had a couple thousand new staff that we hired as well for the new hospital that had to be trained on everything. And then we had staff coming over and even without the new technology, their processes and departments were changing because they had to change. Because they were organized in different ways in the different departments and different locations.
Be realistic about how much can be absorbed. But I think what really worked well for us was we agreed, okay, we are going to have day one and then we’re going to have day two use cases. And we’re going to be reasonable about what we put in day one, but we’ll get the technology foundation established so we could do those day two things. So that’s where we were able to make the heavy investments in a more robust network in the ubiquitous location awareness across the organization in the robots and the automation that we were able to do there and even in some of our patient engagement and other types of technology.
Anthony: Jeremy, that’s wonderful. I think this is going to be very interesting and very useful for your peers when they listen. Thank you so much for your time.
Jeremy: Great, Anthony. Thank you. It was a pleasure.
Share Your Thoughts
You must be logged in to post a comment.