How to Map Out an Effective and Sustainable Remote Patient Monitoring Program
By Eric Wicklund
Healthcare organizations are looking to enhance care management by connecting and collaborating with patients outside the hospital, clinic or doctor’s office, and they’re often doing this with remote patient monitoring (RPM) platforms. Using digital health technology, they’re developing programs that allow a care team to monitor a patient at home, gathering vital signs and other data, communicating with the patient when necessary, and creating treatment plans that can be modified in real-time.
HealthLeaders recently conducted a round-table with three health system executives to talk about their RPM programs and strategies. This panel featured Carrie Stover, MSN, NP-C, national senior director of virtual care for Ascension; Sarah Pletcher, MD, MHCDS, system vice president and executive medical director for strategic innovation at Houston Methodist; and Kathryn King, MD, MHS, associate executive medical director at the Center for Telehealth at the Medical University of South Carolina (MUSC).
What follows is a transcript of that round-table. Quotes have been edited for clarity. This is a two-part feature, the second part of which will be posted on Friday, July 15.
Q. Could you please define remote patient monitoring and how you’ve used it in your health system?
Carrie Stover: We’ve had remote monitoring in place for chronic disease since 2014, but those programs were developed market-by-market and physician practice-by-physician practice – or organically, as I like to say. Historically we defined remote monitoring the same way that [the Centers for Medicare & Medicaid Services] does: Using a connected device and being monitored by a team of nurses.
Over the last few years, and particularly with COVID, we’ve seen that we don’t always need to have a connected device. We’ve started to think about this in terms of a connected care spectrum, from patient nudges, wearables and patient-reported outcomes through traditional remote patient monitoring, and then onto higher acuity monitoring in terms of hospital at home and even some in-facility monitoring.
Sarah Pletcher: It can be any solution where you have a remote care team looking after a patient, informed by some data about the patient. There’s a wide variety – we’ve seen an evolution in how continuous that monitoring might be, or how synchronous, and how robust, and how automated, the depth and scope of data, degree of artificial intelligence, the timeline to intervene [and] what that care team is and can do. And are you using AI and algorithms? There needs to be a continuum, but at the highest level remote patient monitoring can be anytime we are using data to look after a patient, the care following the patient closer to where they are.
At Houston Methodist, we are using virtual care and monitoring technologies to care for patients in hospitals across the health system, as with our VICU program and telesitting service and to offer access to care to the patients wherever they are, as with our Virtual Urgent care, and we use virtual platforms to allow our primary care and specialty care providers to see and monitor patients at home.
Kathryn King: It’s difficult to not have payers define remote patient monitoring for us. Certainly, I think it is helpful in that it gives us a common language. But at the end of the day, this is really about how can we take better care of our patients.
It comes down to, if you could wipe the slate clean from what we were taught in medical school, which is you treat chronic disease in 15 minutes every three months, and instead say, “What’s the one question I could ask my patient every day that would help me better manage their health?” What is the one piece of information we need every day to make a better decision about this patient’s care.
Q: What are the biggest benefits of an RPM program?
Kathryn King: It is about more efficient, effective care for our patients. Whenever you’re talking about a digital transformation or a change in healthcare, that’s what it should come down to: That patients feel like they are getting better access to better care and providers feel like they are delivering higher quality care. Remote patient monitoring is a great example of exactly how to do that.
Sarah Pletcher: It can be much more efficient, allowing us to better use hospital, nurse and doctor resources, and there’s more freedom, convenience, independence, and wellness for the patient, and the technology can allow us to be more proactive. We’re identifying things earlier so that we can impact better outcomes for the patient. Better care delivered more efficiently? that’s the ideal. You’re looking to serve the patient and move the needle forward on the value continuum, so ideally you tick all the boxes.
We’re trying to take increments of time and effort and distribute those for maximum gain. If technology helps me know who I need to spend time [with] in a moment or a day or a five-minute increment, I’d far rather do that than spend it on the people who don’t need me and miss the one that does. If I am the patient, I don’t want to be bothered if everything is OK. I think we’re all trying to get the most bang for our buck.
RPM is the most powerful tool we have available to us, to use data to elevate where we should be leaning in to use more resources, and then being able to pull away and not waste resources that are often disruptive to the patient. If we don’t have to wake that patient up three times a night to check their vitals; but can instead watch them from a wearable patch while they sleep … not only am I not wasting hospital resource time, but I’m also creating a more healing-friendly environment for the patient and catching problems earlier.
Carrie Stover: With traditional benefits, which is why everyone started thinking about remote monitoring years ago, [we’re] reducing readmission rates and improving stickiness to your system. It [also] helps to fill in gaps, what we call the white spaces between physician visits, to help us better understand our patients. There are so many challenges that patients face that we don’t know anything about because they happen in that white space in between visits.
Understanding what’s happening to them clinically, but also socially, is so important. I think we’ve seen some unintended consequences related to improvements in visit adherence by having that relationship with that monitoring team, and decreasing calls to the practice. [We’re] making sure that when we’re escalating something to a physician, that it really requires a physician and that it couldn’t have been handled by somebody else.
Q: What are the biggest challenges to launching an RPM program and how do you address them?
Sarah Pletcher: The challenges are different depending on your RPM journey. At the beginning, it’s ‘What should I do?’, ‘Where’s the ROI?’, and ‘How do I get started?’ Then, [it’s] making sense of a complex and rapidly moving technology market – software, hardware, and considering form factors as well as the potential service partner landscape, and then it’s trying to make sense of the operations.
It often isn’t the fun, sexy things the launch innovative technology but the boring, essential workflows, algorithms, protocols, and change management meetings. The challenges are different depending on where you are on the journey and [they’re] absolutely surmountable, but pack your tool belt before you set out.
Q: Are there challenges that you didn’t expect to have that cropped up?
Sarah Pletcher: Trying to make sense and keep up with the vendor market. I knew that it was complex. I don’t think I fully appreciated all the different niche areas, not of the devices and sensors and gadgets and cameras per se, but all that middle stuff. Data aggregation and data storage and data visualization. And the management software for people using data and how quickly you can build algorithms within the data and when and where EMR integration is essential and where you should avoid it like the plague.
There were a lot more nuances in what I’ll call middleware than I appreciated. It’s been fun to navigate that and come up with a strategy for a very diverse and complex market. But in particular, that middle area required a little more lean-in than I mentally budgeted for.
Kathryn King: Whenever we talk about something disruptive in healthcare, and certainly most telehealth is meant to be such, we talk about culture change. It is a cultural shift, thinking about how you manage patients, particularly large populations. There are a lot of things that go along with that, from provider acceptance to digital literacy. Payment is probably the largest barrier to widespread implementation.
We have seen during the pandemic that when payment and reimbursement restrictions were lifted, a lot of barriers fell away with other modalities. We’re starting to see the same thing with remote patient monitoring.
If we’re only willing to do it within the strict guidelines of what is reimbursed through a fee-for-service architecture, we’re going to limit ourselves and, in some ways, make things less efficient and broaden disparity gaps. For instance, when I ask a physician, “What is one piece of information you need to make a clinical decision?” That might not be a physiologic metric uploaded directly from a device for 16 days during the month, but I have teams that want to make it that because that’s what we will be reimbursed for. That is pushing us in the wrong direction.
Q: Is there one challenge or one barrier or something that came up that really surprised you?
Kathryn King: When I think about this, I think about building programs to increase efficiencies. And I realize that we also need to think about the direct ROI, which looks like dollar signs, but sometimes value can be calculated downstream and in a different way than straight reimbursement.
Getting people on board for that was something I did not anticipate. When we were so excited about these known fee-for-service billing codes, in fact it would threaten to widen disparities by really self-selecting patients, and that would potentially lead us in the wrong direction from an efficiency standpoint. I am happy that we have these codes to put to the test, but I think that what we are learning is that other payment models are needed and we need to keep pushing on that.
Sarah Pletcher: The reimbursement models we have for remote patient monitoring. The billing codes select patients and care teams who can comply with a rigidly defined frequency and cadence of data collection. That’s what worries me, that ‘magical thinking’ that seems to keep cropping up in the RPM space.
People get excited about the technology, and a payment model that can support meeting the needs of the patient, and they think it’s reasonable that a super busy primary care doc whose in-basket is bursting, who’s booked seven months out for just a follow-up visit, somehow they’re going to be able to handle all these incoming alerts, escalations, and alarms and review all this data.
Similarly, for a patient who’s struggling to manage their own care, because they’re elderly or they’ve got comorbidities or dual diagnoses, suddenly we think they’re going to be up for climbing on a scale and attaching their blood pressure cuff and doing several actions every day to deliver enough data.
Then there are the care teams. We forget about the existing burden on our nursing staff and hospital providers and think, ‘Oh, well, it should be no problem for them to do, A, B, C, D, E.’
I find that I’m constantly [playing] whack-a-mole on [that] magical thinking that crops up in the design of these programs. I agree that limited billing codes is one piece of it, and I hope that with more optimized payment examples it’ll be easier to overcome some of these other challenges.
Carrie Stover: One challenge that I was unprepared for was that we have lots of programs that meet every one of the stringent CMS criteria, and it should be so easy to bill for that. It’s 16 days, it’s a connected device, all of those things, but we’ve discovered that we are leaving lots of money on the table because it’s very hard to bill for something in a practice that, number one, is time-based, and number two, is not tied to an encounter.
We have these massive workflow documents and diagrams that detail what we need the remote monitoring team to do and provide to the practice, and then what somebody at the practice has to do in order to turn that into a bill. It’s really complicated. Developing that value, even when you can bill in a fee-for-service world, most of the time we are not because it is so incredibly complicated and complex. It has created so many tasks.
With the program, the challenge is related to logistics and support. As we think about all of these devices and who needs what, I call it the virtual care Clue game. Who’s where with what device being looked at by whom?
With logistics, the best way to enroll someone and get engagement is when they need it. It’s coming from their physician or their discharge nurse in the hospital, and that sounds simple. We’re going to give this patient a kit and they’re going to go home and turn it on. But it’s not simple. Managing lots of locations with lots of variations of those kits is challenging, and sending those kits to the patient’s house, you’ve now created this 24-, 48-, 72-hour gap between when everybody agreed that they need this thing and when they actually get this thing. Are we going to enroll them face to face? Via video? All of those things need to be solved. It’s further evidence that sometimes we, with all of the best intentions, end up with a lot of technology in the closet collecting dust, because we haven’t really thought through the workflow.
With support, you have people who are trying to utilize these Bluetooth connected devices, {but] they may not have the expertise. The vendors all say that everything will work right out of the box, and that may be true for the first two weeks or four weeks or 70 times, but at some point that device is going to come disconnected. How that gets reconnected and who helps them can be an issue. Are they calling the nurses or are they calling the enrollment specialists? If you haven’t really thought it through in the beginning, you have a good chance of having an awesome program fail for all the wrong reasons.
Q: How do you identify patients or the population that you want to monitor at home? And how do you prepare patients for this program? What do you look for in the home setting that will make them good candidates for an RPM program?
Carrie Stover: That’s been part of our journey from 1.0 RPM to 2.0 RPM. 1.0 was a little bit of a free for all. We’d take anybody who wants to participate. In 2.0, we’re more thoughtful and specific about inclusion and exclusion criteria. How do we identify patients that are appropriate clinically? How do we identify people who are at the right trajectory in their disease progression to participate? Then, understanding who’s going to benefit the most from these programs from a clinical perspective, because we can’t monitor everybody all the time.
When we think about assessing a patient from their ability or interest in participating, that engagement is so much better when that discussion is initiated by their physician. There’s no substitute for that conversation.
We also think about the home environment. Even though these devices are cellularly connected, we have regions where patients have zero cell coverage. That’s part of our assessment. We have a team of enrollment specialists to determine whether a patient should be kitted or unkitted, whether they can use an app or an unconnected device vs. a connected device or a tablet. They assess their interest in the program, their willingness to participate, and help walk through that enrollment process.
It creates a scenario where we’re [eliminating] the patients who may need our help the most and may have the worst access to other forms of care. We’ve been really specific about creating equity solutions, but it doesn’t solve the problem by any stretch of the imagination.
Kathryn King: From a population-based approach, at one end we’re talking about high-touch, technologically based remote physiologic monitoring, which poses a lot of barriers. At the other end, where folks are starting to explore, is the lower-touch, not-as-specific, technology-based monitoring of patient-reported outcomes – asynchronous engagement and therapeutic monitoring. That’s what we’re tending towards from a standpoint of equity: What is the lowest tech solution to reach our clinical goals.
At that point you’re looking at what population can I engage in this way. We’re piloting a lot of programs. One that has been particularly successful is in perinatal behavioral health, [targeting] anxiety, depression, and substance use during pregnancy, because we know those are tied to the highest causes of maternal mortality in our country. That tends to be a time when people are very engaged in their healthcare. We’re trying a lot of programs, both for mothers and newborn babies, that look like a low-tech, text message-based monitoring programs.
If we’re debating the right population for this, we need to ask ourselves if this is a population that we need to engage in a different way. They’re not a great population for X, Y, or Z. Well, then maybe we need to do A, B, or C to better monitor and engage this population.
Sarah Pletcher: It’s trying to fit the program to the population that you’re trying to reach, and then choosing the most minimalistic and successful tools to achieve the goals that you’re aiming for. For sending a patient home with a kit, for example, you need the ‘just plug-it-in’ level of simplicity, whether that’s a cell or a WiFi, or even just using their own device. And in terms of sensors, it can be as simple as a sticker that they wear home from the hospital or a whole suite or a toolkit when truly needed.
One thing that is underconsidered is it doesn’t always have to be the patient managing these tools themselves in their home. It can also be leveraging a partnership. Are they getting visiting nursing? Is there a family member who can assist? For patients who may not have cell or WiFi connectivity, are there places that they can go, [such as a] church, grocery store, community or senior center, soup kitchen, that can be data hubs, that can be a partner if the patient’s going there anyway? Can those be our distribution centers rather than always trying to make it work in the patient’s home environment?
Selection and enrollment are critical, but if there’s a population that we’re struggling to reach, we’ve got to think outside the box to bring that program to them, even if it means redefining where that is.
Q: How do you develop support among your staff for these programs? What are their most common concerns about RPM, and how do you address them?
Kathryn King: When you start the conversation, wipe the slate clean. Why do we think we need to manage things in 15-minute increments every three months? Most providers are really on board [and] know what they would ask their patient every day. Where we run into some hiccups is in the algorithmic management of patients by a centralized team of nurses. We hear from providers who say, ‘Well, I really manage my patients individually, and there are certain things that I need to consider on an individual basis for each one of my patients.’
We direct the conversation toward gold standard treatment guidelines, and we want to work with them to develop an algorithm that will help them adhere to the highest quality guidelines for the majority of the population that really fits into that algorithm. Once we have an algorithm that they feel comfortable with, we will use it to take care of a large percentage of the population in a way that they would want them taken care of. There will be some patients who do not fall into those guidelines, and it will be very clear through remote patient monitoring that those patients need to be escalated to the provider for individualized treatment that only they can provide.
Basically, you frame it as ‘How about I take all of the patients that are really easy for you to manage and make sure the patients that you really need to see get to you.’ Usually they say, ‘Oh yeah, that does make sense. That would make my life easier.’ Again, we’re not taking patients away from their provider; we are helping to [identify] the patients that they have to spend a larger amount of time with.
Sarah Pletcher: We have a buy-in matrix for providers with three core pillars. They are time, money, and tech. It can’t be a huge waste of the provider’s time or an intrusion into their off-time. It can’t cost the provider or have them lose out on reimbursement to participate. It doesn’t have to be a huge windfall or save them loads of time, but can’t be a big loser on either of those dimensions. In terms of the tech, it has to be basically decent. It doesn’t need to be earth-shattering. It just needs to be reliable and do what it’s meant to do with modest support.
If you get those things, then providers can be motivated for lots of different reasons- maybe to look after a chronic condition by value metrics, or because it’s cool or because they get to be the first person in their practice to try something or to create convenience and peace of mind for their surgical patients. There are all sorts of ways you can amplify buy-in, but I haven’t seen a program succeed if it doesn’t have solid foundation in those first three pillars.
Carrie Stover: With providers, that has been a challenge. It reminds me of the early days of the EMR when people thought there was no way that we could ever come up with consistent order sets by disease process because every physician treated their patients differently and individually. We have spent a lot of time and energy undoing some of the organic growth of remote monitoring.
With RPM, historically there have been centralized monitoring nurses whose job was to take that alert and pass it on to the physician. We’ve spent some time thinking about the job of that remote monitoring nurse. Just as a patient in the hospital is evaluated by a nurse before anyone calls a physician, we do the same thing. And as we start to monitor higher acuity patients, we have a team of very highly skilled nurses who have lots of clinical experience. We’ve developed very clear assessment protocols and escalation paths.
We’re finding that 60% to 70% of the time, nurses can resolve that alert. The rest of the time they may be initiating a patient case and sending a note to the physician. Very rarely, about 1.5% of the time, they are actually escalating to a physician. And then that’s resulting either in a medication change, a virtual visit, or perhaps a trip to the emergency room.
The message that we’re creating for physicians to get buy-in is that we’re not looking to inundate you with additional data points and information. We’re trying to create a valuable set of information for you to better interact with and treat your patient population.
Eric Wicklund is the Innovation and Technology Editor for HealthLeaders.