Tracy Rausch is the co-founder of DocBox, a company developing an advanced clinical process management solution for hospitals. In this episode, Tracy explains how device connectivity can eliminate medical errors and improve clinical workflow. We’ll also discuss why data centricity is similar to the thought process of a physician and what the future of the healthcare industry will bring.
In Episode 20 of The Connext Podcast:
- [0:33] - What was the motivation to start DocBox?
- [4:51] - How did DocBox discover Connext DDS?
- [10:05] - Are there competitors currently occupying this space?
- [14:36] - Interoperability between medical devices
- [21:38] - How do you see the next 5 to 10 years in regards to global adoption of this technology?
- [29:32] - What is the pre-submission process like for new medical products?
Related Content:
- [Blog] Using a Data-Centric Approach to Building Healthcare IIoT Solutions
- [Customer Snapshot] DocBox: Innovative Clinical Process Management Solutions to Improve Medical Safety
- [Datasheet] RTI in Healthcare
- [Whitepaper] A Data-Centric Approach to Developing Digital Health Solutions within the IIoT
Podcast Transcription:
Steven Onzo : Hi, and welcome to episode 20 of the Connext podcast. In today's episode, RTI's healthcare expert David Niewolny interviews Tracy Rausch. Tracy is co-founder of DocBox, an innovative company developing an advanced clinical process management solution for hospitals. We'll discuss the importance of interoperability, and how it's fundamental when integrating medical devices for scale. Tracy will also share her perspective on what the global healthcare industry will look like in the next 5-10 years. We hope you enjoy this episode.
[0:33] - What was your motivation to start DocBox?
David Niewolny: Why did you start DocBox? What was the defining moment, and how did you actually take that and turn it into something tangible?
Tracy Rausch: So, yeah. So I started DocBox as I was working as a clinical systems engineer, and the problems that were arising, I couldn't find a technical solution to actually meet those problems. And as I moved through this and met with vendors and went through and started saying these are my requests, as my customer, I still couldn't get the solutions we needed. And so DocBox actually came out of that frustration of trying to work with vendors to make that work.
David Niewolny: The fact that you have this overall problem, that there's a single person in the industry that does not see it from the medical vendors, from the healthcare providers, to the insurers. And from that aspect, from what I can tell, you're the only person doing anything about it. Would you say that's accurate?
Tracy Rausch: I wouldn't say I was the only one doing about that. There's a pretty big team of folks on the back end of this who are working on this from a research perspective, on the safety critical piece of it and the cybersecurity piece of it. And the other aspects of it, DocBox has just been the fortunate beneficiary of about $30-40 million worth of funding of the R&D side of it, that we've moved it to that commercialization phase.
David Niewolny: And really, when you're talking about that research, that's all the work that Dr. Julian Goldman and the medical device plug and play group out of Mass. General and Harvard has really contributed.
Tracy Rausch: Right.
David Niewolny: So you kind of use that as a baseline...
Tracy Rausch: That's the base. Kansas State University's done a lot of work. University of Pennsylvania's done a lot of work. There are some other ... the other groups and entities around that have actually been part of this. There's also large chunks of government employed researchers who are actually working on this too, at the FDA and the Army, other aspects of it. So we're not the only piece, we're just the top of the triangle right now on what this infrastructure should look like, so.
David Niewolny: So would you say of all that research ... Again, being one of the only commercially available products today that is out here really taking this problem head-on, taking the best of each of these, do you call it a standard? To me, when you have a company moving forward with a product, it's much more of a product, because you're not a consortia or a research organization or a group that's actually trying to put together an open standard. But at the same time, what you have very well could be.
Tracy Rausch: I think the goal is that this does become a standard, or something that looks like this. I think once it goes out into the wild, it's going to change again. We've never claimed that we've got this right to start with. We've got that foundational piece. There's a lot of stuff that has to be added. So once we put this out there, then people can actually start to work on the hard problems.
David Niewolny: Okay, interesting. One of the things that I have grasped is DDS is core to, overall, the offering that you put together. You mentioned during your talk today, you were talking about the ... I don't know if you call it the data bus. I know we call it the layered data bus, but basically, you showed the same sort of layered data bus model that we have, and it definitely seems like that's core to your product. Can you talk a little bit about how that's being implemented?
Tracy Rausch: Yeah, so we generate this, what we call the ICE bus, that's one domain for every patient, is how we look at that and how we move that going forward. And then if you want to scale, you just add more ICEs to your system and your environment. So that's really been that foundation, and one of the big reasons why it's so important is you, as any individual component of the system, doesn't know what your information's going to be used for. And you don't know who's going to need your information. So you have to have a many to many communication relationship. You can't have a peer to peer relationship. So that's why that bus is a foundational piece, is that if you're a medical device, you don't know what the intended use of that data's going to be once it leaves your system. And you can't predict that.
Tracy Rausch: And that's been one of the challenges of being able to implement this stuff, is that traditional, you've had to know what it was intended to be used for. And the technology's moved way past that, and we just have to catch up.
[4:51] - How did DocBox discover Connext DDS?
David Niewolny: That's fantastic to hear. It's definitely something that ... I think DDS, as a core technology, has been used in a lot of other industries. Correct me if I'm wrong, but I'm assuming you tasked your team to go out and take a look at all of the available technologies that could do this many to many communication for this secure, interoperable framework. Is it accurate to say, you guys looked at everything that was readily available, and DDS was really what rose to the top?
Tracy Rausch: Yeah, and the way that it happened was I told my software guy, "Look. Here's the requirements, go find me the piece that works." And after about two weeks, he came back and said ... And you have to remember, this is 2011 ... he came back and said, "Hey, there's this thing called DDS. Are you interested in looking at it?" And I said, "Okay, let's go through it." And I said, "Okay, this is interesting."
Tracy Rausch: The piece that drew me to it initially was the ability for quality of service, because we know we had to go to being able to do control. And so I knew that I was going to have to have certain pieces of the system that were best effort, whereas another piece of the system was going to have to have that harder timing issues of what's going on. And that was really the start of it. I will say in 2011, we were pretty naïve of what the system was going to look like at the time, I think, as anything you would develop going forward.
Tracy Rausch: But that was that key piece of it, that ... really what drew me to it, but then there was all these other things. I think something interesting also is the data-centric piece of it is really how physicians think. And it's very counterintuitive to most engineers.
David Niewolny: They all think database, database, database?
Tracy Rausch: Well, they think message. Structured message, data, back and forth, whereas a doctor says, "I need a heart rate and a blood pressure and a respiratory ..." So it's actually somewhat intuitive, if you spend the time on the physician's side and the clinician's side, they think in parameters, they don't think in device message world. So that data-centric piece, it wasn't something at the front of our mind, but later on, I kind of put that together and said, "Wait a second. This is actually the same, thinking of how they solve problems and how they look at stuff."
David Niewolny: I guess in terms of the technology, I think we all see the fact that you have interoperable, secure, essentially, doctor in a box that can now align all of these multiple vendors, benefit to the overall industry. What do you see those benefits are? I know you've done some homework on that.
Tracy Rausch: Yeah, so I think the biggest thing is that doctors have to be able to solve their own problems. They're going to need answers to these solutions. There's no team of engineers that's going to be able to go into a clinic or understand medicine in a way that someone who's spent 8, 10, 12 years of medical school and 20 years of experience ... That's the information we want to get into the box. It's not what's in your medical school books. It's when a doctor walk in the room and knows what's wrong, it's that intuition piece that you've gotta get in the box.
Tracy Rausch: And so the doctors are going to be the ones who are able to solve that problem, and so we have to enable them to innovate. And right now, they've been locked out of their data, they've been locked out of their information, so it's freeing up this information and this data for them to do the work. And I think a lot of people don't realize that they've been doing this work since the '80s. And they've all hit a roadblock, because they don't have good quality, synchronized, contextually aware data.
David Niewolny: Coming from that clinical environment, seeing, standing next to the doctors, working with them, you've actually seen that physical pain. I think myself from the outside, you look at this and you start seeing more or less the high level facts of their leading cause of death being medical error. And I think you showed a couple examples of real medical error happening in a clinical environment. And you also mentioned today just the amount of time that clinicians spend entering data, and I can understand, that's not only frustrating for a clinician, but also if I was running that hospital, I'd be really frustrated that the highest paid people on my staff are spending all this time essentially entering data. So there's gotta be a huge cost savings, as well as just from a humanity aspect, saving lives is huge as well.
Tracy Rausch: Right, and I think it goes back to, the doctors and the nurses are innovative people in and of themselves. And they're going to innovate their way ... I mean, they innovate their way out of workarounds and problems now. Give them the right tools to innovate to fix their own problems, and they're going to go do it. And so that's really where the focus of this has been, is that there's this large community of researchers who are already doing real time decision support. AI and machine learning is not new to the medical field. I think it was called neural networks before. That's the story of where it went.
Tracy Rausch: They've picked this stuff up. As soon as it comes out, the first piece of it, you know that they're doing it. I think most people would be surprised that there's a large chunk of doctors who all know how to write code. They moved down that pathway. That's what's going to move the industry forward, is the fact that you've got two or three different medical institutions all saying, this is the newest algorithm to do this. This is the best way to do this. This is the new gold standard. This is the data. But the other piece of it is then they have to be able to scale that solution, and that scaled solution comes with an infrastructure that lets them do that.
[10:05] - Are there competitors currently occupying this space?
David Niewolny: And that's really what you see yourself providing, is that baseline infrastructure. All of that said, I think this market definitely has the ability to grow and be significant. If you're really selling a box per hospital room, you're looking at the number of hospital beds, hospital rooms, really globally. It's exceptionally large. So do you see any other competitors in your space? Do you expect competitors to play in this space?
Tracy Rausch: I think the best way to answer it is, I hope there's competitors who play in that space, because that means adoption has actually started to occur, and I think you're going to see that innovation. You're going to see a new market of software as medical devices, and the ability to do that and deploy that ... If every single algorithm developer, every single person, has to worry about building their infrastructure first, it takes off the innovation piece.
Tracy Rausch: 40-50% of the grant money is just to get the data. This is a challenge that they're all dealing with, and I want to emphasize, it's not just your electronic medical record data. There's context and there's physiological waveforms and there's all this other information that comes together to be this really big story of medical data that's available.
David Niewolny: And that's really what you're providing, is the medical data. And then from that, you're hoping that the ecosystem develops. I guess to kind of reword that question, do you see any competitors in that space in terms of trying to capture that medical data?
Tracy Rausch: I think there's some folks that are capturing it. That's not the interesting piece of this. They've all been able to capture the medical data. It's knowing how to scale it and deploy it and make it useful, is what the biggest key piece of this comes into play.
Tracy Rausch: And we're talking about different pieces of medical data, too. There's your billing codes and your billing data and your reimbursement, and your actuarial and your statistics, and the stuff that's publicly available from CMS and all that stuff. That's one set of data. And then there's this other set of data of what's actually about the patient, it's what happening. And then the third set of data is the emerging set of the genomics and proteomics.
David Niewolny: Yeah.
Tracy Rausch: And the three of these actually are the complete data set. We're not saying we're the complete data set. All three of these tell the story of that patient's history through their care, sells them the story of their care. That's the complete medical record of that patient. It's not the physiological data, it's not the billing data. It's a combination of the two. What is that story, how did they get to provide care, because there's research questions that happen all the way through those steps. Why did they make that decision to give that drug, or why did they make that decision to do this? Was it an insurance decision, was it a medical decision, was it an availability decision?
Tracy Rausch: We don't know the answers to these questions, and it's going to be really hard to study what's the best care if you don't have the whole story. And that's the pieces, we're trying to add our piece of that whole story. But the thing is is that no one of these can be proprietary. They have to be open and they have to be available, because the answer is going to come from not one entity. It's going to come from the community actually innovating.
Tracy Rausch: You need all three of these pieces of data. There's not competing, we're not trying to take over one or the other. This is just a gap that's sitting in that patient record. And this data may or may not belong in the traditional electric medical record. Five days' worth of EKG waveform is probably not useful to the history of that patient, but in certain instances, it might be. There's no physician who's going to get a patient record and it's four or five terabytes of data, and they're going to go through minute by minute, the EKG. But they may learn from that data, and that's the ... How is it being used, where is going, what does it look for, is really the key.
David Niewolny: This is a standard IT problem. It's like, gather all the data. Once you have all the data, you look at it, figure out what to do with it. And we're kind of now in that really early phase. With the IoT, we're starting to gather data, but we still have all these proprietary systems, and we really don't have that full story. So once we get the full story and all of the data to play with, now we can actually start making sense of it. But what you're saying is we're really not even there yet, or we're starting to crack the surface.
Tracy Rausch: Yeah, we're not even close to being there yet. There's so much we don't know, that you can't know what your data's going to be used for, because if this data starts to become available, you're going to see this rapid increase in innovation in healthcare. And there really hasn't been disruption in healthcare since the mid '80s. The mid '80s into the early '90s, there hasn't been any earth-shattering, groundbreaking new piece of technology that's hit the market that's changed the way medicine's practiced.
[14:36] - Interoperability between medical devices
David Niewolny: I've been involved in technology in healthcare now for right around the same amount of time, 15 years, and I thought the holy grail was connected medical devices. So connecting them via ethernet, connecting them via wifi, and then you had wireless sensors. And honestly, it's kind of what brought me to RTI, was you got the point where you found out connectivity was really just the start. You start connecting these devices, and you can now capture the data, but you still don't have full datasets. You don't have that interoperability.
Tracy Rausch: Interoperability has morphed into a marketing term, and it hasn't morphed into what it actually means. And in healthcare, it's a safety issue. And it's a data analysis issue. And it's another piece of data. I can take all this data, I've done it since the beginning of my career. I've taken all this data and thrown it in a database, and you can do that for 10 patients. And you can come up with a new algorithm and you can analyze it. That's been going on for the beginning of time for medicine. But how do you take that solution and scale it over 5,000 beds, 10,000 beds?
David Niewolny: Yeah, it's physically impossible.
Tracy Rausch: That's where the interoperability comes into play, and that you have to be able to do this at scale, because that's how you iterate on that algorithm. And right now, an algorithm is 7-10 years to come into the market, because of not having interoperability. And by the time it gets to market, there's already four other ones in literature that are better than that algorithm, and that's the ...
Tracy Rausch: So if you look back at that and you think about that as ... That's how fast researchers are innovating. Our technology's not keeping up with them. And we've gotta give the hospital the tools to be able to swap out to a new algorithm. And do it the way ... oddly, do it the way they've always done it. Peer reviewed, best practices, gold standard innovation. And to move forward to do the next thing, because it's not a widget and a technology, it's not a consumer product, that you're going to buy the next generation in two years. This is ... Medical devices have a life cycle of five to seven years, and you're dealing with safety critical, mission critical pieces of technology, and you have to treat that with that respect.
Tracy Rausch: And I think the biggest thing is, I always remind people, is you are a patient. So think about, would you want this connected to you? Would you want this moving forward? And I think that's the biggest key. When you start to think of it that way, you take a little bit of a step back and say, okay, we need to make sure we have this correct.
David Niewolny: Definitely seems like the forefront of a real big inflection point within the healthcare industry. It seems like you're kind of leading that charge. I guess looking at the old guard, which is the medical device companies, how do they view DocBox?
Tracy Rausch: We have relationships with a lot of them, and they actually want to work with us, because hey have the same problem. I think people don't realize that no vendor makes everything, so they always want somebody else's data. And it's really hard for two competitive vendors to say, "I'm going to share data." But it's really easy if they're both talking to a third mutual entity to say, "I'm going to give you both the same data that's available."
Tracy Rausch: So they have intellectual property. They have algorithms that they can't get into the market, because they're running into the same problem as everybody else. Your little vendors have a bigger challenge than your larger vendors, but they all have the same problem when you get down to it, is that they're landlocked with their own information. But that's not the full contextual picture of what's going on, and so they can't get to where they want to go, because they don't have enough information to actually move to the next level.
David Niewolny: Interesting. It sounds like you're definitely solving a problem for them as well.
Tracy Rausch: I hope so. And as I said, this is a community. This is to bring all of the innovation to the table. There's a lot of great algorithms out there, there's a lot of ... It's there, it's available, it's been developed over the last 20 or 30 years. And we need to move that into the market to advance it.
David Niewolny: A lot of what you've mentioned to me, it's exceptionally inspirational. Someone coming from a clinical perspective, seeing the problem firsthand, being frustrated enough with it to go out, tackle it, solve it, but what you have created is a business. So I guess it's one thing we haven't hit on, is business model. In terms of, who are truly your end customers? We talked about hospitals, we just talked about potentially medical device companies having access to that data. What are the type of business models that you see DocBox moving into, or the industry moving into?
Tracy Rausch: Our major customer is the provider. That's the key. How do we provide the best technology available to solve their problems, and how do we allow them to have that and actually grow with it? We know this market's not fixed. We know this market's going to change very rapidly. Even if new algorithms were every three years instead of every seven years, that would be a huge acceleration, which ... I know I'm sitting in Silicon Valley, the tech world is laughing at me at this point when I say that. We're used to a six month tech turnover cycle. I'm talking about a 7-10 year tech turnover cycle right now. Just shortening that is a big deal to go forward.
Tracy Rausch: So I think we have that piece of it. I think we also have this untapped value of data in healthcare to solve medical problems, to solve efficiency problems, to solve operational problems. And those problems all have their own different business models and markets as they go forward and they move to do this. And I think the really cool thing is that it's really the same set of data that solves all three of those problems.
Tracy Rausch: So ideally, we're that infrastructure and that platform. We're an app developer, and we're going to become a data company. And that's the three pieces of this that we'll move through, and that's been the strategy for the company to go forward, is those three aspects of it.
David Niewolny: Well, it definitely seems like a way to make an impact across many different areas of an industry that definitely could use the help.
Tracy Rausch: Yeah, and the passion behind this is, I was on that other side. I think my heart and soul is still on the provider side. And so the two worlds don't have to be in conflict with each other, is what they currently are. They actually can work in parallel and partnership. It's a relationship between the two, it's bringing that engineering and that technology and meet those clinicians halfway to solve their problems instead of the forcing of saying, "Hey, here's a new widget, here's a new tool. Here's a new piece." It's, no, let's understand this from a systems standpoint, and put the solutions in place with the right technology to go forward into it.
Tracy Rausch: And I think it's going to change. There's going to be new sensors, there's going to be new devices. There's going to be new visualizations. We haven't changed our central station visualization since inception. We haven't changed our user interfaces of devices since inception. This is the mid '80s. This is what I'm saying, that innovation has not occurred in this industry since the mid '80s to early '90s. And that's a long cycle for any sort of disruption to come into the market.
[21:38] - How do you see the next 5 to 10 years in regards to global adoption of this technology?
David Niewolny: So I guess looking into your crystal ball, how do you see the next 5-10 years? You talk about refresh cycles that are going, say, seven to ... up to 15 years. Do you see that moving much more quickly now? Do you see DocBox ... Obviously you're starting to gain some traction. How do you see adoption globally?
Tracy Rausch: So I think adoption globally is going to call up a couple places. One, the emerging markets are going to leapfrog the non-emerging markets, because they're green pastures. And so they're moving in with, what is the right technology, what is the place? You're adding hundreds of thousands of beds a year in these markets. They have a lack of staff, they don't have a lack of patients. They're technology savvy enough. The world is evolving that way, so they can adopt those new pieces and those solutions. So that's going to be one area that is important.
Tracy Rausch: And I think the crystal ball is, there's five billion people without healthcare in the world. We need to bring healthcare to them. That's the altruistic goal of this and how you look at it, but that infrastructure and that platform, the hardware's going to evolve to whatever the need is for. But we use off the shelf components, so we don't have custom PCs, we don't have ... Everything is off the shelf, so we can rapidly switch that technology, which is one of the things that hasn't been able to be done in the past.
Tracy Rausch: And as you add applications, you completely change the platform. So if I'm in a cardiac ICU and you go to a medical surgery ICU, there's a completely different set of apps. We can change that out without actually changing any of the hardware. This moves this capital intensive, hardware intensive industry to software by shared resources and shared components and has their information, as you look and go forward to do that.
Tracy Rausch: It's all the same DocBox infrastructure, and then the hospital can pick and choose the apps. So they can pick the best in breed. Let them pick the best algorithm on the market. Let them install two different algorithms and run them head to head and test them and understand what's going on. Let them pick the best sensors and test them. And when we do this with our cell phones, we do this with our appliances, we do this with everything else that we buy. Why would we not do this for the most important thing that you have to do, which is taking care of your health and your life? We can't do that right now. We can't put these head to head tests and say, no, this sensor's actually better than that sensor. There's no way to actually do that right now.
David Niewolny: All of that definitely makes sense. I think one of the things is healthcare is one of the most highly regulated industries out there, so getting to three, four, six months refresh cycles, I think, could be quite challenging. I think one of the things that I've seen, the FDA has now put in place a digital health unit. They've been in place now for a couple years. I know Dr. Goldman has talked or worked or consulted with them on both cybersecurity and, I think, in some forms of interoperability. Do you see that as being kind of a key driver for improving these refresh cycles?
Tracy Rausch: I think it's going to be a key driver, but I think we have to ... The regulatory science has always been part of this work. It's been part of this work from day one. We're engineering in the safety instead of using it as an afterthought, and that's one of the things where the consumer digital health world has ... The FDA has always been the barrier to entry, and that's really not the case. It's build good software, do it under a quality system. And it does, it takes longer, there's more testing required. But that's not the seven year issue. Companies need to re-look at, what's the regulatory pathway, what's the science behind this, how do you get a safe and effective system, 'cause that's really down to the key, is how do you build a safe and effective clinical system? Not so much, well, I gotta get through the FDA.
Tracy Rausch: And when you look at it from that perspective, then you add additional functional requirements to the system. But the platform has been designed with that aspect in mind. DocBox's commercial entity is based on several years of regulatory science, research that's occurred through academic researchers and other things. And how do you actually build a system to do that? And how do you educate the community on how you actually get this going forward?
Tracy Rausch: And that's one of the key things that ... We've been federally funded, so that's part of that work. What's the regulatory science behind that? But the regulatory science actually started before the platform design in DocBox.
David Niewolny: Okay.
Tracy Rausch: Which is a different approach. But it's not bad, it's just, you have to think about this piece because everybody says, well, it's just a sensor. What can that impact? Well, where does that sensor data go? What decision's being made with that sensor data? Are you sure that data's correct? Those start to become ... And then everybody's like, yeah, yeah, absolutely that needs to be ... There's a regulatory piece to that. And I'm like, well, you're part of that ecosystem, so you have to be a good actor.
Tracy Rausch: I think people compare it to, say, well, healthcare and patient safety or aviation. It's the same thing. Or it's manufacturing process, and you've gotta do the engineering to where it fits into healthcare. So many people want to force fit this stuff into healthcare to say, well, it should just translate, it's a one to one. And it's not a one to one. There is an art to medicine, and you have to respect that there's still an art to medicine.
Tracy Rausch: And part of that comes from the optimistic side of me, is there's an art to medicine because we don't know everything yet. And therefore there's this art piece of it, and that art piece is going to be there forever. We're never going to understand everything, and so-
David Niewolny: Until you're able to tie that genomic data plus all the patient data, all of that pieces-
Tracy Rausch: But then there's this other humanity approach to this, of this interaction of the patient, and how does the environment impact how you care, and this whole other ... nurture piece of this, and how that impacts care-
David Niewolny: It's Eastern medicine meets Western medicine.
Tracy Rausch: Right, right. We just don't know the answers to all of these questions, so you have to respect that there's this art, and the technology has to meet that art somewhere in the middle. The science is going to have to work in parallel with this, because it's a complex system we're analyzing, and we're not at the end yet, and we may never be at the end yet. And so it's not physics, it's not chemistry. It's a combination of all of it. It's electrical signals, it's chemistry, it's physical components. It's mechanics. There's all these aspects, they're all thrown together. We're a really complex system, and so we have to have some respect.
Tracy Rausch: And there's no blueprint. We do not come with manuals, we do not come ... So we're reverse engineering the system. And so we're not going to understand all the pieces quite right yet.
David Niewolny: Mind you, we're all built different.
Tracy Rausch: Yeah, your physiological data is probably pretty darn close. That's the key. You're running on a system. So we're reverse engineering a system that we're trying to understand to be able to act on that system, and we have to have that respect for that, this industry and the entity that it has. So yes, all this other technology can come in from all these other industries, and we should. We should be adopting it. And you're bringing it into the culture, you're bringing it into the art, you're bringing it into the thousands of years of history of medicine, which ... I don't think any other technology industry actually has that long of a history of what medicine has.
Tracy Rausch: And so it's an art, science, technology, and you've gotta balance the three.
David Niewolny: It's really a different look. Again, I think it's where it puts you in such a great position, working side by side and seeing that, rather than having outsiders from outside the industry come in and try to solve your problems.
Tracy Rausch: Yeah, clinicians aren't very good at expressing their requirements. They don't think that way. They don't think linearly, they don't think ... A doesn't solve B, which is C, because they're dealing the human body. So A plus B can be two, four, and seven, depending on what the situation ... And they learn from practice, they learn from repetition, they learn from understanding how they learn, how the clinician actually operates. What does a clinician actually do is really the key, it's not just understanding the science behind the disease or the practice.
Tracy Rausch: And that's the two pieces that ... You have to meet somewhere in the middle of the two.
[29:32] - What is the pre-submission process like for new medical products?
David Niewolny: In my experience, when you're taking a product through the FDA, it generally needs to be a complete system. Hardware, lower level drivers, software with the application. So it's basically solving a problem, has a bunch of data on top of it. So when we're talking about moving to a much more app-based world, and potentially, you could have different pieces of hardware that this is living on, do you have any idea how the FDA's going to handle that? Or is that another challenge that we need to overcome?
Tracy Rausch: I think part of that's been a misnomer. The FDA's accepted off the shelf computing devices now for a decade. So you set requirements and you say, these are the minimum requirements I need. You show that those requirements are met, and they're absolutely ... that's normal. The mobile medical app guidance actually is the first piece of guidance that actually decoupled these two. If you actually look at it says. Because it's saying mobile, people assume that means your phone.
Tracy Rausch: And there's been substantial work. Dr. Goldman actually worked with [MDIS 00:30:34] and others for a community-wide [de novo 00:30:37] submission after an FDA meeting in 2010 about device interoperability, that they had a meeting. They went through a de novo filing of a platform-based medical device to actually do that. So that process was actually done with the community, several vendors involved, several academic institutions involved. They went through that filing process with the FDA as a research project. They filed pre-submissions, they had formal meetings, they got the response. It's all publicly available information.
Tracy Rausch: They're looking at risk, they're looking at efficacy. You can follow what you need to do for any of those. There's standards that help you, there's other things. But it's really how you want to prove to them that you have that risk. I always encourage people to use the pre-submission process with the FDA. You can go in, you can ask them questions, you don't have to have everything done before you go in and have that conversation. But you have to be open to understanding their side of the piece, because their side of the piece is, how do I make sure this is safe for the public, and how do I make sure this is effective? That's their goal, and that's the pathway they go down.
Tracy Rausch: And so they have the ability to question, how are you going to make sure this is how this works? Listen to what they have to tell you. And you may not like what they have to tell you, but they're going to tell you what you should do and how you should go forward.
Tracy Rausch: And we've been through three pre-submissions with the FDA. We've went in and said, is this what you meant? Is this what you wanted to look at? This is how this is going to do this, this is our approach to doing this going forward. Open and transparent, which has not always been the case. People lock up and don't want to talk to them, and as I said, our platform is open, and that means the regulatory piece of this is open too.
Tracy Rausch: The infrastructure, it's a big value add, but it's not the secret sauce. The secret sauce comes into all the different applications.
David Niewolny: I'd say for any of our listeners out there, there's some pretty solid advice in terms of any software products, apps, or even hardware devices going to market, going through that pre-submission process sounds exceptionally useful.
Tracy Rausch: And they've just announced that they're going to put a pre-certification process in place. They've got their pilot companies, and they're moving in that direction. I think everybody understands that medical devices are becoming software, if they're not already software. People may not think they're moving fast enough or they're not doing this, but this is a really complicated problem, and I think you hear so frequently, "Well, they didn't let us do X. And I don't understand why they didn't let us do Y," until you turn around and say, "Would you put this on your child? Would you put this on your family member? Would you wear this yourself? Would you depend on this information?" And that's where all of a sudden you see, okay, wait a second, this is a safety critical system. We need to do this correctly.
Tracy Rausch: There's enough in history where this didn't go correctly that has hurt or injured folks or people, and there's a reason why this stuff is in place.
David Niewolny: Same thing we were talking about with autonomous vehicles. We're trying to really put a system in place that is better than any driver out there. And I think we're really looking to do the exact same thing with healthcare and medical, and this is really just ... DocBox provides that enabling technology to make that happen.
Tracy Rausch: Right. And the people who are going to make that happen are actually the clinicians themselves. They have to translate this knowledge. It's not in a database somewhere. It's not in a medical school book. It's literally them.
David Niewolny: The best cardiac surgeons in the world are going to be coming back with new apps that are going to help better track certain cardiac issues.
Tracy Rausch: Right. How do they take this, how do they deduce their reasons, how do they do their differential diagnosis? And that's the key piece of this, is we can't do it any other way, because that's been medicine for the last 2,000 years. Doctors train other doctors who train other doctors who train other doctors.
Tracy Rausch: And the amount of information that a doctor with 30 years' experience, what they had in medical school is not the same thing. But how you actually diagnose and provide care, it's by repetition and it's handed down from physician to physician, and that's the key piece of this. And there's always that physician that says, "We can do this better. Why don't we try X?" And then they hand it down to their people and their folks and that becomes the change and standard. But it's an iterative and generational process, that we're trying to extract that information to understand, what are they seeing?
Tracy Rausch: And I think the key is that it's not just the doctors. You have a nurse who actually is the one who uses our system more than anybody else. It's the nurse who's doing the documentation. It's the nurse that's doing the heavy lifting. Carry the orders, caring for that patient, and do those things. But how is that nurse who walks into shift and can tell you which patient she's going to have problems with? There is a reasoning behind that.
Tracy Rausch: So it's not just decoupling and understanding the doctor. It's also the nurse. And understanding what decisions they make, what are they doing, when do they know there's an issue? 'Cause you have a skilled ICU nurse, she can tell you everything. When you walk in the room, she can just look at you and know, you know what? There's something wrong. Now, why can she look at you and tell you there's something wrong? There's nothing that tells her that besides experience.
Tracy Rausch: And how do you actually figure out what that is? It's not about going and talking to that nurse and saying, "How did you do that?"
David Niewolny: If you want to break it down to ones and zeroes, you're right. Her brain is looking at that, and she's like, "I have seen this so many times, and when I see it, this is what's happening."
Tracy Rausch: And so the key is, that means there's data there. And so we've gotta get that data. She's operating on something. But we just don't know what that combination is, and that's what we have to find.
David Niewolny: And the exciting thing is I think that's just the first phase, because I think what's so cool about getting access to all this data is once you have access to it, you'll start seeing interactions that the human mind can't even comprehend. When you start-
Tracy Rausch: Right, but they may do it subconsciously, because of experience. You just don't know why you leap from A to B. You just know you've done it before. We've gotta figure out why you leap from A to B. That's the long term future of this, is enabling this infrastructure that can innovate, can move these apps, can generate new applications, can generate new ways to do this. So when you see that doctor right out of medical school, versus that doctor that's been done 30 years ago, they do things different.
Tracy Rausch: It's not that ... Some of them are better than the guys doing it 30 years ago. He's got more information. But they don't think and program and read the exact same way. So we've gotta understand those differentiators too, of that young physician looks at things differently than the older physician. And who's right? We don't know. We don't have any data to know which one's right anyway.
Tracy Rausch: So that's the piece of this that we have to go through and look at, is based on their background of information, how they figure things out. And that's what this data is going to be able to do. But it all comes down to being able to collect the data in a way that you can actually analyze it and look at it and leverage it to start to figure this stuff out.
David Niewolny: And that's exactly what you and DocBox are doing.
Tracy Rausch: That's the goal.
David Niewolny: Fantastic.
Steven Onzo: Thanks for listening to this episode of the Connext podcast. Stay tuned for our next episode, where we talk to RTI's principal applications engineer about benchmarking the performance of Connext DDS versus open source DDS.
Steven Onzo: If you have any questions or suggestions for future interviews, please be sure to reach out to us either on social media or at podcast@rti.com. Thanks and have a great day.