“Even the small hospitals really do want to enable a modern consumer web experience today”
In this episode, Graham Gardner discusses how his company, Kyruus, targets patient access issues in health systems. Inspired by the Moneyball concept, Kyruus uses advanced analytics to understand patient needs and put the right providers “to bat” where they are most likely to do well, thereby reducing or eliminating variations in the quality of care.
Graham discusses the challenges of introducing digital innovation in healthcare and provides some counterintuitive advice to other digital health startups looking to overcome challenges in sales cycles and achieving enterprise-wide adoption for their platforms.
Welcome to The Big Unlock podcast where we discuss data analytics and emerging technologies in healthcare. Here are some of the most innovative thinkers in healthcare information technology talk about the digital transformation of healthcare and how they are driving change in their organizations.
Paddy: Hello, everyone, and welcome back to my podcast. This is Paddy and it is my great privilege and honor to have as my special guest today, Graham Gardner, CEO and Founder of Kyruus. Graham, thank you so much for joining us and welcome to the show.
Graham Gardner: My pleasure Paddy. Thanks for having me.
Paddy: So, let’s get started, Graham. Would you like to tell us briefly about your company, its evolution, and what is the marketplace need that Kyruus is trying to address?
Graham Gardner: So, Kyruus is a routing and scheduling platform for health systems that helps to ensure that patients within their network are getting to the right network provider. And the founding vision for the company was originally Moneyball. And this idea that one can use statistics to understand relative competencies amongst players and use that information to put them up to bat in situations where they’re more likely to do well. And the team, therefore, benefits from everybody playing to their strengths. And I was convinced that you know, we as physicians, the same way as ballplayers trying to hit every pitch. We as physicians, we’re trying to help every patient. And yet, statistically, we had huge variations in the care, and cost, and quality that we were delivering to patients and began to wonder whether we could get a more granular view of providers, their statistics, if you will, and use that to understand what they might be best suited to treat and put them up to bat, if you will, or match them with the right kinds of patients. And as we began to engage with more and more healthcare systems around that conversation, what we uncovered was that many of them were really struggling with the ability to match supply and demand scale. And what a lot of them would acknowledge to us was, we know patients are waiting far too long to get in and access care here. As a result, they’re either not getting that care or they’re leaking out of the network and seeking care elsewhere. What was staggering to me, though, Paddy, was when they would next say, it’s not because we’re 100 percent full. In fact, 20, 30 in some cases 40 percent of the appointments were actually empty every single day. And the reason for that was that too often patients were queuing up for Doctor Famous. You know that many of the other physicians are trained with someone at the call centers had on the shortlist to send to. But meanwhile, there might be a brand-new doctor who did exactly the same thing as Dr. Famous, and she was sitting at 30 percent productivity. So, as you kind of looked across this large network, you had a lot of demand not being accommodated, a lot of supply going underutilized. And it was only getting worse because health systems were getting larger and larger so no one really kind of knew each other anymore. At a time when medicine was getting super, super subspecialized. And so, gone were the days of my go-to ortho guy that I send all my orthopedic cases to. There was now a hand person, an elbow person, , foot and ankle, spine, et cetera. And people really didn’t have the right information to understand where to refer a particular case, that was right for the patient, right for the provider and even at the top of their system and also right for the healthcare system. Paddy, we looked into it, in healthcare you don’t have to invent anything new, you have to look at whatever else was in the 1980s and bring it forward. And so, we actually looked at the airline industry, which had actually wrestled within many ways solved a lot of the same seat utilization challenges. Turned out if you rewind back to 1990 every time an airplane took off in this country, about 4000 seats were empty. So very similar to where some of these health officials found themselves today. And what had transformed that industry were things like the Saber system that ultimately became the likes of Travelocity and Kayak, it depicts the idea that on one screen you can look across all the providers, you could look across the American Airlines or Continental or JetBlue, in our case, doctors, that might be on the Epic scheduling system, the Cerner or Athena, you could sort and filter by business logic. So, if I wanted to take off after 9:30 and sort by price, I could click and do that quickly. In our case, if I felt more comfortable as a female provider that spoke a certain language, I could again click and find who as soon as available and basically book in and get that seat, if you will, on a single platform. And so that really became the model for how we thought we could help the health system achieve operational, financial objectives. But most importantly, get the right clinical care to each of its patients. So today we work out about 600 hospitals around the country, about 250,000 providers on the platform and continue to enjoy partnering with health systems and increasingly partnering with other companies as well that really kind of that overall patient access solution.
Paddy: Fascinating. How long have you been around?
Graham Gardner: I’ve been around for about 9 years.
Paddy: Nine years. So, you know, I really like the Moneyball analogy that you made. And that, of course, is very catchy and intriguing at the same time. And of course, you’re addressing a problem that is all too well-known, which is patient access, which comes in many shapes and forms in terms of its impact on a particular health system. The utilization rate is obviously one which is very, very visible because it has immediate revenue implications for the health system. And I’m very familiar with this problem. So, let me ask you this: when you start working with health systems, how are they addressing this problem today? And can you give us an example of how you used, let’s say, a proprietary algorithm that you developed to help them get better at what they do? Can you maybe illustrate with a simple example?
Graham Gardner: Yeah. So, it’s very much an industry in transition. I think when we first started, to be perfectly frank, organizations were working on paper and, you know, go into call centers and you would see an agent desk, cluttered with different kinds of paper. There might be a binder that had, profiles in our directory that have been printed out, a year ago and was likely updated the minute it was printed. There might be special instructions scribbled down with someone’s back-office number there as a posted note. You know, Dr. Jones doesn’t take patients on Yankee game days and has a poor scheduler, who is very much the front line and the first voice into a healthcare system, is having to look in all these different disparate datasets to try to route someone to the care. And I think that really was so much of what accounted for the inefficiency and being able to kind of make that match. The second thing you still see today, Paddy, is that different parts of the organization we’re working with different solutions. And so, you know, there might be still a print directory going out to all the referring providers. And that was where people were getting their information there. But the marketing team that was running the website might be compiling a different set of information and they might be doing outbound phone calls to the practice to try to keep the address current. Or they might be sending out surveys to the doctors to try to do that. Then the call center might have a completely different set of information that might includecredentialing. And too often with that different information, the poor patients were seeing real inconsistencies across that. They might get a name from their referring provider. They go online to do some research and actually they’re looking at a different specialty or something like that. Then they call into a call center, book the appointment and find out the person is not even working there anymore. And so, again, there’s a real lack of standards and the right tools to do that. I think people began to look in different places for solutions. Many then turned logically, they do to their EMRs, what they could learn there. But EMRs are really not built to do that kind of keyword searching and routing and certainly not across different EMRs as well. So, I think it’s for that reason, Paddy, that a lot of them begin to look to us to see what we could put across this different EMR as a platform to kind of enable that that routing.
Paddy: So, I want to kind of unpack a couple of things you said. As far as the provider matching is concerned, there are two aspects to the problem. One is the quality of the information as it relates to the provider data and the other part is actually finding intelligent ways to direct the patient traffic in ways that help a patient get what they need. And also helps distribute the workload in a way that the overall utilization level goes up. Are you addressing both problems or just one of them?
Graham Gardner: Yeah. Great question Paddy. We are, although I would say I’ve been surprised in our journey how much of it is still today, the basic data management. So yes, when we first began talking with health systems, our hope was that we could use a lot of data to uncover who is most efficient or most effective in different areas. And what we heard from a lot of health systems early on was, look, we just want to figure out who our doctors are, and we can simply get knee cases for the knee surgeons and hand cases to the hand surgeons. You know, that’s a huge lift for us. And so, a lot of the work that we do today, Paddy, is really compiling very robust profiles of providers. We do that by tapping into dozens of different systems. We bring around 200 different data elements to bear in a profile. So, the data element might be a personal statement, it might be a specialty training location, it might be a back-office number, things like that. And so, we pull those from different systems. And a lot of the work that we do is not only integrating the data. So, for example, I might be Graham S. Gardner, M.D. there and Dr. Graham Gardner over here and, you know, scale making sure those are the two same identities. But actually more important is really the governance around that data. So, for example, as I was alluding to earlier, a location might be pulled in from a credentialing file, but that only gets updated every two years when someone’s re-credentials. It might be in the marketing database, so they’ve made outbound phone calls every six months to verify where someone is, but it’s also in the senior practice management system where the provider seeing patient’s tomorrow morning. That’s probably the most accurate one. And so, we do a lot of that work early on and engaging with the health system to understand where to pull the information and then set up all those feeds to make sure it’s current and governed and sign off around it. That’s the first part. But then where we next go, Paddy is enriching that data with the clinical library of terms, conditions, procedures, diseases, and things like that. We over the years, beginning with claims data analysis that then really curated with a lot of clinical experts and now our clients as well. It’s been a developing 25000 term library of conditions and procedures that we crosswalk to the different sub subspecialty. So, when we onboard an organization and identify, eight peripheral interventional cardiologists, they get assigned anywhere from a hundred or 250 terms to describe the typical practice of a peripheral interventionist. Oftentimes that provider, whether their practice manager is actually engaged in that process and actually further curates and can actually do at the condition leveltell us what they want to see in their practice. And that, Paddy, is a huge physician engagement piece to what we do because I think for providers who are working really hard and suffering a lot of burnout, the idea of a robot coming in and filling up their panel and getting it wrong, is understandably really scary thing for them. And so being able to kind of extend them a tool that says, look, we actually understand exactly the kind of patients you want to see and we’re not going to be responsible for sending you a knee case if you’re actually a hip surgeon. So that it is really kind of invites them to be a part of that process. And the first step of beginning to kind of ensure that quality is being accounted for. Well, you’re getting people to the right people for that care. We take additional data sources, for example, pristine scores and things like that so you can begin. First of all, you can publish that to patients, but also potentially, rank how you distribute care based on those scores as well. So, we’re beginning to take more and more steps into quality metrics and things like that so we can ensure that the data is right, but also that the organization begins to operationalize the strategy for how it wants to distribute patients within the network.
Paddy: Right. So, who is your buyer for a solution like this? Chief Marketing Officer, Patient Experience Officer, who is your first protocol for something like this?
Graham Gardner: Yes, Paddy, this has really been one of the challenges. But I think something that we’ve embraced and do much more effectively today. So, the reality is, as I described, that of a multi-channel approach to scheduling and patient access, there are a lot of leaders in the organization thinking about this and looking for solutions. And so, we’ve gotten very comfortable talking with a number of different stakeholders and quite frankly, can vary by organization, Paddy. So, you know, one example, it might be the organization decides to centralize their call center. They brought 200 agents onto the same room. And it’s really the, head of access to the COO that’s now looking for a technology platform to help enable those agents to be as effective as possible. Another organization that might be Chief Marketing Officer who is, competing with the organization down the street that’s now enabling online booking. And she now wants to enable that for her organization. And so, it starts off much more like a digital front door type of conversation. In others, it’s around trying to stitch together clinically great network and its clinical leadership that’s beginning to look at that. So, we oftentimes will start a conversation in various parts of an organization. What we do as part of that conversation is to try to bring those leaders together. And even though we do sell our package in modular ways, we have a call center package, web site package, and then a more of a distributed point of care package. Organizations can buy one or more of those. But we do like to engage all those leaders in the conversation right upfront to help them understand the power of standardizing on one platform across all that. So, all those leaders, obviously the CIO and the IT infrastructure has to be engaged. We’re tapping into many of their different systems. And ultimately the CFO is often the one writing the check and really seeing and appreciating the benefits of taking an organization that may only be at 60 percent utilization of its schedule and deriving that 70 percent. And, you know, a multibillion-dollar organization has a profound financial impact.
Paddy: Yes. Well, I’ve been in healthcare tech a long time and one thing that we all know about the provider ecosystem is that the sales cycles are very long, and I just think from what you just said that adding a layer of multiple stakeholders that you need to convince. Does that make it longer? Does it make it more challenging and difficult for someone like you?
Graham Gardner: It does. It certainly makes it longer, without question. And yet, Paddy, I think what we feel is that it positions the company very well because no other company has really taken the same approach to incredibly robust data management. And then that multi-channel approach to matching across all the different access points for an organization. And so, it really kind of positions as well as the single vendor that can really provide that solution for folks.
Paddy: You may have answered this question already, but in terms of ROI do you demonstrate it in a very straightforward way by just showing higher utilization rates or how do your stakeholders make a business case to invest in a platform like this?
Graham Gardner: Yeah. That’s been a very important part of our conversations with clients. I’m a cardiologist by training Paddy. And so, I grew up in the world of clinical studies. And, you know, from day one, we instrumented the tool to really begin to understand the impact of what we were doing. We think about it in different areas. So, one is really around patient acquisition, and this is, I think, easiest to demonstrate on the digital channel. You know, when you take an organization live and now enable patients to book online. We see tens of thousands of patients within a month start to book online. And what’s incredibly exciting for these organizations that oftentimes half of those patients are brand new to the system. So, they were you know, they were shopping, and consumers are starting to vote their feet. And if someone allows them to book online, they will select that at 8:00 at night and book in. So that was very exciting for them. Eighty percent of the patients are commercial, by the way. So, if, as you know, of course, I have to think about the business side of medicine, you know that’s obviously exciting for a client. But the other aspect of it, which I had not picked up on early, but talking with our clients about how much they appreciated this year, that’s 10000 phone calls, Paddy, not going to a college. And so, from an operational burden perspective, you’re actually acquiring those patients without staffing up a call center and things like that. That’s been exciting for people on the kind of acquisition side of things. What we see on the retention is a 50 percent reduction in leakage, 50. And this is always something that I believe would be able to demonstrate when I first started learning about this industry and hearing terms of leakage or patient outmigration. You know, I was convinced, Paddy, that the patients aren’t referred out of the system because doctors are bad people trying to sabotage the system. They honestly just don’t know who their own colleagues are. And, you know, their go-to person is now Dr. Famous who can’t see people for months. That patient needs care. And so oftentimes, yes, they will leak out of the network. And what we found is that if you do provide a tool now for people to find their colleagues and they will do the right thing and refer in-network and really keep that care coordinated with the same EMR systems and all the rest of it, so that’s been exciting to see. In the last, we just really kind of home run metric is going back fill the schedules. And so, what’s exciting is we will take a baseline of the schedule a year before we go and then measure that again a year later. So, you can get a comparison of the flu season. And sure enough, we increase schedule densities anywhere from 2 to about 11 percent, which doesn’t sound huge. But then again, when you talk about a multi-billion-dollar organization, if you’re bringing in, 10 percent more patients at the top of the funnel, it generates tremendous amount of downstream radiology and labs and procedures and things like that. It’s been very exciting to build it and demonstrate that with customers.
Paddy: Yeah, I know. That’s a nice segue to my other question that I was going to come to. You mentioned these are multibillion-dollar organizations. And we talk about digital front doors and clearly that’s making a huge difference in terms of access. But is there a certain profile to whom this kind of a solution is more applicable and more beneficial than others, for instance? You know, I don’t know that community hospital in a city environment. I don’t know. I’m just asking, is this more applicable to them versus maybe a big city hospital in Boston? Is there a sweet spot? Is that a kind of an ideal sort of a health system that can get the most out of your solution?
Graham Gardner: Yeah. So, you know, nine years ago, Paddy, my mind thinks exactly where yours were, which was, you know, that the larger and the more academic an organization, the more I could utilize this. Because you know that there really is a left nostril surgeon who doesn’t like the right nostril surgery and, be able to kind of really find that needle in the haystack was going to be somewhere where we could shine and allow people to really filter down by a number of different things. It’s also where you’re getting a lot of complexity in multiple different systems for the benefit of a platform sitting on top of that and standardizing everything was going to be really beneficial. And so, for the first several years that really had been what we targeted based the top 400 health systems or so in the country. Oftentimes they’re in very competitive metropolitan areas. And so, again, that they really are looking for something that can differentiate them from what I would say, though. So, over the years, we’ve begun to go down to smaller and smaller organizations that I think particularly around the digital front door. Paddy, I think, you know, in the call center, look at there really only are two orthopedic surgeons. Everyone kind of knows who does what in a small hospital that’s maybe less compelling. But everybody, even the small hospitals right now really do want to enable a modern consumer web experience. And to the extent that involves rich data and videos and the ability to book directly online, that is something we provide. And as we’ve developed our own solution and now scaled and operationalized the way that we can, you know, onboard an organization that size more effectively, we’ve been able to keep going in that direction. So, we haven’t found anyone too small for us yet. We certainly don’t do small physician groups, so we don’t do the 18-person orthopedic group or anything like that. But we do a wide range of healthcare systems.
Paddy: That’s interesting. So, let’s take a step back and look at the broader picture here. Now, you are part of the digital health ecosystem of billions and billions of venture government money that’s pouring into innovative startups. Some of them have done very well, others are struggling. Based on your own experience, can you talk about what are some of the biggest challenges that you have had to overcome? You’ve been around nine years, so that is long enough for me. So, talk to us about a couple or two or three of your top challenges that you’ve had in your journey so far.
Graham Gardner: Yeah, honestly surviving is one of the key things these organizations move very slowly, particularly when you’re trying to introduce a new market concept or segment that we were very lucky early on to find wonderful champions that could take us to organizations. But there really is a lot of alignment that has to happen. And, you know, we watched our own space, mature over the years, from something that one or two people saw. You have the likes of Aaron Martin, such a forward-looking thought leader in this space to this really becoming a CEO level initiative that now, the C suite was responding to and had a budget to go out and fund. And so, I think part of it is surviving and raise the venture capital support, which has been terrific to allow us this to kind of live long enough, quite frankly if I wonder the market wakes up. I think that’s one part of it. I think the second part is really having a humble approach to the industry. I think that maybe less so today. But certainly, when we were starting, there was a lot of feeling that healthcare people are behind, know they’re 30 years behind everybody, they don’t know how to do things. We can come in and just solve things from a tech perspective in a much more efficient way. And the reality is that healthcare is different. You know, I will use the analogy of airline seats for what we do. It’s finding the right provider is more difficult than booking a seat of Chicago on American Airlines flights. And so I think coming in with that humility and really listening and understanding, these are people who care deeply about patients and not making mistakes and, you know, people to sell to hard [00:32:04] snake oil salesmen. I think this industry is trained to repel that. And so, it really you have to bring credibility and listening skills, I think, in order to kind of understand the challenges. And then I think the third thing, and this is something that we’ve figured out over time, Paddy, is how to be the right size. And what I mean by that is that so much of what I see, you know, even getting funded today are point solutions and end up in the pilot world. And if you don’t have broad organizational buy-in, it’s very, very hard to scale something like that, and it’s so easy these days to spend 10,15, 20 thousand dollars on something but really not get buy-in. And so, there’s a certain surface area that I think you need to speak to in order to be noticed, quite frankly. And you know what we’ve noticed, Paddy, over the years as our deal size has gotten larger, it commands more attention from the organization and those end up being much more successful implementation than customers because there really is that buy-in across the organization. So, I think being big enough is one thing. But then I think on the other extreme, Paddy, it’s not trying to boil the ocean and do everything. And so, a lot of where I’ve been spending my time over the last year or so has been in forging a number of business development partnerships with other vendors and technologies that serve the healthcare system. So, you know, the example we announced the deal with Salesforce, and you have an embedded turnkey solution now within their CRM because the reality is that health systems are looking at both scheduling but also CRM. And so, we’re doing more of these kinds of partnerships with things in telemedicine, then chat technologies and things like that so that we play well with others and could be part of a Lego set if you will, that all kind of works well together.
Paddy: Yeah. So, you compete in some area as your partner and others. Is that kind of the mantra?
Graham Gardner: That’s right. And you know, a lot of the deals are just understanding the boundaries. So, hey, I’m going to go up here. You’re going to do it from there. And how do we work together and serve the customer the best?
Paddy: Yeah. So, you know, you mentioned the pilot plant. Actually, some of the things you mentioned a little bit counterintuitive if I may say so because you would think that the small deals have an easier chance of entry because, you know, it’s not a big financial commitment and it’s easier for people to make that kind of commitment at least from a buyer standpoint. But what is counterintuitive also is that larger deals get more attention and also are more likely to get to organizational buying and support to ensure the success of the program. That’s kind of a little bit counterintuitive, but it’s very insightful as well. Let’s talk about the pilot thing. You know, death by pilot, pilotitis, you know whatever flavor of that. That seems to be the pain of the startup ecosystem in healthcare. In addition to long sales cycles and all of that stuff, what is your recommendation to health systems to break through this and, there is a mutual win-win here if both parties can get together and pull out a solution, that is where there is an obvious ROI? But it doesn’t seem to be happening fast enough. So, what is your recommendation? What is your wish list, if you will, for health systems to be doing in order to accelerate the adoption of innovation?
Graham Gardner: Yeah, I know. It’s a great question and what I think a lot about Paddy. I think on the health systems side you know what I enjoy the most is where an organization really has identified its priorities. And, you know, I understand. I am sure you go to HIMSS every year. And I remember standing in that football field of vendors one year and realizing, oh, my gosh, this is what the inbox looks like for someone at these health systems. I’m population health, I’m population, pay attention to me. And so, I can understand that it’s overwhelming the inbound demand that comes in. And so, I think where organizations have clearly articulated these are the four or five initiatives. You know, this is how we’re thinking ourselves proactively about stitching something together. Those tend to be organizations that have really done the work and understand what they’re looking for and make, you know, faster, and better decisions. So, I think that’s one of the things that’s really taking a proactive view as opposed to the kind of a reactive one. But then I think when there is a priority go in on it and, you know, there’s obviously a time and we did a handful of pilots very early on, Paddy when you’re just trying to figure out the layout, design, and understand this. But the quicker than you can move to say no, look, this is something that matters to us and we’re going to invest in it, and we’re going to invest in time, we’re to invest in money. And, you know, a couple of examples for us Paddy, they’ve actually invested the money as well in terms of equity. So, you know, a couple of our health systems are actually investors and encourages as well. And yeah, that has been a really wonderful partnership. Not that we’re ever going to return a meaningful investment to a 10-billion-dollar healthcare organization. But for the individuals that are there and putting sweat equity and introducing us and caring and trying to make a successful feeling that they’ve got skin in the game and are part of our success, I think has meant a lot to them as well. So, I think if that investment loses time and money picking something and saying, look, this is something we’re going to really work hard to try to make successful because it can’t just be on the vendor. We need too much from the organization to be successful. And so, I think it’s been committed to that.
Paddy: Fantastic. Well, Graham, it’s been such a pleasure speaking with you and thank you for setting aside the time to talk to me. I’m sure my listeners are going to have a lot to chew on from all of your comments and all the very best to Kyruus, all the very best to you and your team and all success to you. Thank you again.
Graham Gardner: Thank you. I really, really enjoyed the conversation and look forward to talking more.
About our guest
Graham is the Co-Founder and CEO of Kyruus where he has led the development and commercialization of the company’s market-leading patient access platform that now serves over 250,000 providers and 500 hospitals. Prior to Kyruus, Graham was a Venture Executive at Highland Capital Partners where he co-founded Generation Health, a genetic benefit management company that facilitates optimal utilization of genetic testing, and served as the company’s Chief Medical Officer through its acquisition by CVS Caremark.
Graham completed his clinical training in internal medicine and cardiology at Beth Israel Deaconess Medical Center and Harvard Medical School, where he also served as Chief Medical Resident. Graham completed his BA and MD degrees at Brown University and earned an MBA from Harvard Business School. He serves as an advisor to Sigma Surgical, Sensory Cloud, and the Innovation & Digital Health Accelerator at Boston Children’s Hospital.
About the host