Interview with Dr. Adam Robison – Episode 333
In today's episode, Dr. Adam Robison explains how he was able to found a start-up, AI Medica, while practicing hospital medicine full-time.
This interview will reveal the impact of AI Medica's software on healthcare efficiency, its integration with Electronic Health Records (EHRs), and its role in enhancing clinical decision-making. Adam provides his firsthand account of navigating the complexities of healthcare technology while practicing medicine.
Our Show Sponsor
We're proud to have the University of Tennessee Physician Executive MBA Program, offered by the Haslam College of Business, as the sponsor of this podcast.
The UT PEMBA is the longest-running, and most highly respected physician-only MBA in the country. It has over 700 graduates. And, the program only takes one year to complete.
By joining the UT Physician Executive MBA, you will develop the business and management skills you need to find a career that you love. To find out more, contact Dr. Kate Atchley’s office at (865) 974-6526 or go to nonclinicalphysicians.com/physicianmba.
Did you know that you can sponsor the Physician Nonclinical Careers Podcast? As a sponsor, you will reach thousands of physicians with each episode to sell your products and services or to build your following. For a modest fee, your message will be heard on the podcast and will continue to reach new listeners for years after it is released. The message will also appear on the website with over 8,000 monthly visits and in our email newsletter and social media posts. To learn more, contact us at john.jurica.md@gmail.com and include SPONSOR in the Subject Line.
Revolutionizing Healthcare Efficiency with AI Medica
During our interview, the founder of AI Medica discusses the journey of his software optimization company and its groundbreaking role in transforming healthcare data accessibility. With a focus on integrating with Electronic Health Records (EHRs), AI Medica streamlines medical calculations, provides coding review, and enhances clinical decision-making while using your EHR.
Navigating Entrepreneurship in Healthcare Technology
Adam also delves into the challenges of founding a company while maintaining a thriving clinical career. From overcoming the limitations of existing EHR systems to securing investments and building a business, he shares insights from his entrepreneurial journey. During our conversation, he highlighted the major steps he followed:
- Identifying a universal problem,
- Leveraging personal experience,
- Describing the technological solution,
- Collaborating and investing,
- Navigating the complexities,
- Continuous learning, and
- Balancing clinical practice and entrepreneurship.
These steps resulted in the creation of AI Medica, a company poised to streamline healthcare data accessibility and decision-making. Doing so, helps physicians work more efficiently and improve quality of care.
Summary
To learn more about AI Medica and connect with Dr. Adam Robison, you can visit the official AI Medica website. For inquiries and demonstrations, you can contact Adam directly via email at adam@aimedica.ai. Additionally, you can reach out to him on LinkedIn for further information.
AI Medica offers a revolutionary solution to streamline healthcare data accessibility within Electronic Health Records (EHRs), making clinical workflows more efficient for physicians. If you're interested in exploring how AI Medica's tools can benefit your healthcare institution or clinical practice, feel free to reach out and schedule a demonstration.
NOTE: Look below for a transcript of today's episode.
EXCLUSIVE: Get a daily dose of inspiration, information, news, training opportunities, and amusing stories by CLICKING HERE.
Links for Today's Episode:
- AI Medica website
- Apply Discovery, Innovation, Value, and Execution to Launch Your Business – 319
- Here is the Designer of the Best Medical Image Repository Ever – 125
- Top Five Start-Up Secrets for the Physician Entrepreneur – 087
- How Blockchain Technology Can Protect Your Credentials with Dr. Leah Houston – 045
Download This Episode:
Right Click Here and “Save As” to download this podcast episode to your computer.
Podcast Editing & Production Services are provided by Oscar Hamilton
Transcription PNC Podcast Episode 333
How to Practice Hospital Medicine and Found a Startup
- Interview with Dr. Adam Robison
John: I'm really happy to meet today's guest because besides being a hardworking hospitalist, he's a hospital manager and leader and founder of an EHR optimization startup company, which we'll talk about for sure. Hello, Dr. Adam Robison. Thanks for coming today.
Dr. Adam Robison: Thank you so much for having me.
John: I am very happy that you're here. This is going to be very interesting. People know that I have an affinity for hospital management work. I was a CMO of a hospital and I know a lot of physicians are always trying to get away from it, but I always like to talk to people like yourself who are in hospital leadership positions and of course, also an entrepreneur. So this is going to be fantastic.
Let's just start by you telling us a little bit about your background, education and clinical career, and then we'll get into the other stuff after that.
Dr. Adam Robison: Yeah. My name is Adam Robison, like you said. I am an internal medicine doctor. I trained at the University of Louisville. I'm a big Cards fan. We did pretty good this year in football, so I was pretty happy about that. I've been working as a hospitalist clinically for the past seven and a half, almost eight years now. I work in a small community hospital out in the middle of nowhere in Idaho, in Twin Falls, Idaho. It's a great place to practice and I've been out here for a while and we are here for quite some time.
On top of that, I do work as a lead hospitalist for my group. There's about 120 providers and we cover about four or five sites right now. And so, that's been interesting work. I took that role all about almost three years ago, right at the middle kind of beginning stages of COVID. That was a very interesting time to take over as a leader of a large hospitalist group and try to navigate through that. That was good learning experience.
John: Yeah. I'm tempted to say, "Well, okay, how did you solve all those problems of people being too sick to work and not having enough PPE?" But we won't get into that. But it's been interesting, huh?
Dr. Adam Robison: Yeah, it's been interesting work. And then yeah, as you said I did found a software optimization company about five years ago, almost five years ago, 2019. And so, that's been a lot of work too. It's been a lot of interesting stuff that keeps me busy and a lot of different things to focus my attention on for sure.
John: Well, it's that second part that really got me interested. And I think our listeners are interested in things like outside work, side gig, side jobs, new careers, passive income, active income. Tell me what inspired you to start a new company? We'll see if your story jives with the others I've heard in terms of what makes people do crazy things like that.
Dr. Adam Robison: Yeah. I've been happily married for almost 20 years, and my wife, I asked her one day this crazy idea. I said, "Hey, I want to take some money. And I have this idea of how to make the EHR better." And she said, we'll have at it. And so, basically I remember sitting in front of my computer, I was using Epic at the time, using an electronic health record and going "I have to go to a third party website to finish some work. I have to go outside the EHR on a regular basis." I'm like "I have to take data from here and go kind of chart and go over here. This seems kind of stupid. Is there a better way of doing this?"
And so, initially I tried to do what I did initially within the EHR with some templates and stuff like that. And what I was looking to do wasn't actually possible within the EHR framework, the logic and the kind of advanced computing I wanted done. And so, that's what caused me to think about is there a way of doing this? And I did a lot of research. I was reading all sorts of websites, educating myself on interoperability standards. How does that work? Is there ways of doing this outside the EHR without really tight integration customized integration? And then that's what we came across. I came across something called HL7, which if you're out there is fire. And then we learned a lot about that. Really what was interesting to me was a problem that I had clinically and could I solve it. And that's what led me to found the company, realizing that we are founding a company to develop the software and to go from there. So, that's what we did.
John: That is the common thread that I've heard before. I talked to an urologist who started a company producing underwear for patients, and whatever, other different entrepreneurs. And it's always that itch that has to be scratched. You have a problem, but you're on your own, and it's like there's got to be a solution to this problem. That is very consistent. But I'm sure the way you went about it is going to be different from others because there's a myriad number of ways of doing that. We'll get into what it does exactly, but what was your next step? You had this idea, what did you do?
Dr. Adam Robison: Yeah. Whenever you have any idea, we're trying to find is there a market out there for it? Is this something that people would buy? Is this a universal or at least a broad enough problem that people run into it? I talked to lots of different physicians and people I knew, people at different hospitals I had connections with and asked them "Is this a similar problem you run into, or am I just dumb and nobody else has it?" And see if somebody else has already had a good solution for that. And realize, no, this is a fairly universal problem that everybody else has experienced. There's not a lot of solutions out there in the market. And so, that's when I go, okay. Now I don't know how to code or software. I don't have any experience with that. And I knew it would probably take me quite some time to do that. That's when I was looking around for people that knew how to do that and I was able to find some developers that had some of those skills. So, it's a lot.
There's stories about how to get licensing and getting approval through different EHR vendors. That was an interesting experience to go through, to try to convince a certain ethic that I was a company and that I wanted a license to put myself on there. Learning how to navigate all that, security protocols, yada, yada, yada. Just a lot of learning new things that I didn't know before.
John: Interesting. I just want to make sure I didn't miss another point. You checked around to see if this was a universal problem, and did that include other EHRs and the one that you were using?
Dr. Adam Robison: Yeah. I talked to a bunch of physicians because of what I have learned about, and it sounds like you talked to other doctors that have used this, but physicians are the best people at solving physician problems. Not some sort of a Silicon Valley startup by two guys out of their twenties. Like "Oh, they know how to fix our problems for us." It needs to be physicians fixing physician problems. And so, I talked to a bunch of other doctors that have been in a myriad of different EHRs. I talked to people that used the big ones in the market as well as like the VA. A bunch of colleagues worked at the VA. Is this something that you'd experienced out there? And it was a pretty universal experience.
John: Tell us about the company. Let's start there. Tell us what the company is designed to do, or what kind of product or service it provides. And then I'll come back to some of these little questions in here.
Dr. Adam Robison: Yeah. The problem that we're trying to solve was how to get data out of the EHR. It was the problem. That's a big issue. Healthcare data and how it's siloed right now is a big problem. And that's why I touched on briefly something called the HL7, which is a group, they develop these interoperability standards. It was very serendipitous timing with what I was trying to do, because during that same time, the 21st Century Cures Act had come out. And the 21st Century Cures Act for those that are familiar opens up the chart. For a lot of physicians, that meant that patients now could access certain notes in real time, which has got a lot of physicians from heartburn. All of a sudden the chart became a lot more. It used to be My Note, now it became the patient's note as well as My Note. And that was some heartburn around that.
But if you actually looked into what the 21st Century Cures Act was doing, it wasn't just making notes available to our patients, even though that's how we allowed, as physicians interpreted, it actually made the data in the EHR transparent and accessible. It had to be accessible. And it turns out, the government had adopted these things called the HL7, the standard called FHIR. And it just happened the same time while that was going on that I was trying to look to solve the same problem of how to pull data out of the EHR and do something with it meaningfully.
And so, what our tool does is it integrates directly with the patient's chart. It looks at what's going on with the patient's chart, and then pulls all the information that would be relevant to pull out, reviews the chart essentially for the user, and then provides information like MELD scores and things of that nature as well as coding recommendations all within the context of the patient's chart.
John: Interesting. Because earlier I was going to jump on the bandwagon and bad mouth some of the EHRs and EMRs. It's like you would think after helping physicians doing this for 10 or 15 years, they would've figured this out. But what you're telling me was the system wasn't necessarily ready for it and it became ready as this was implemented.
Dr. Adam Robison: Yeah. It's been interesting. I will tell a story. I have a good friend of mine who's not a physician, and him and his wife, they recently had a child and they were receiving care locally to smaller hospitals. But the child's medical condition required it to go to seek care at a tertiary care center. And what they described, even though both centers used the same EHR, they're separate EHR instances. And so, the mother and the child had two separate records that were actually very important, but then needed to talk to each other because the conditions they were seeking care for were the same at both sites. And this is just recently.
And so, for whatever reason, there hasn't been a lot of interest in the major EHR vendors to make even within their own EHR network or broadly between EHRs, that hasn't really been an interest that they've wanted to go down either maybe for economic reasons or technical reasons, for whatever reason that may be. That interoperability standard piece has been a big problem. And they haven't been willing to fix it and the Cures Act did mandate they had to open that up. But it's going to be important on companies like mine and other companies out there to do that work for them.
John: Okay. That's good. That things are progressing, and like you say, your timing was perfect. Now, I was looking at your website and reading a little bit about this, so I just want to make it clear for the listeners because some of them might actually need to use this tool at some point or tell their hospital or somebody about it. There are different kinds of metrics, there's algorithms, there's sometimes formulas that have to be used by a clinician and trying to do something. And right now you have to either do that by hand or use a separate piece of software or an app on your phone or something, and you're integrating that. So, give us more about how that works and why it's potentially a profound app.
Dr. Adam Robison: Yeah. Our two applications right now, we have two software pieces called Aicalc, Aicode that live within our Aimedica platform. And what they do is the Aicalc, you can imagine it just being your medical calculator, your cirrhosis scores, your chads bask, you name it. There's a million of these out. And they're actually propagating pretty rapidly now. With the advent of big data and smart computers, these models are coming out pretty rapidly. And our software basically looks at the patient's context, what kind of medical problems they have, is there other criteria, and says, oh, not only we'll calculate, oh, patient's cirrhosis, you'll need a MELD score. And then we'll give you the most up to date MELD score, give you that sort of stuff. You don't have to click around. You just click the button. You don't have to think of why you need the MELD score. We'll just review the patient chart and give you all the relevant scores and pull the data in to calculate the scores for you automatically.
Again, one thing I say to people all the time, I use this tool every day in my practice. I developed it for myself, and so I'm constantly trying to improve it. And then we also do the same thing with medical diagnosis codes. We try to optimize the best highest weighted codes for DRG waiting for hospitalizations, which is important because we can look at a patient's chart and say, "Oh, the patient actually meets sepsis criteria and you are treating of infections. That should be the code you put in the chart. Let's provide that." And that has a huge ramification. And anybody knows in the healthcare space, those DRG waiting is huge for hospitalizations reimbursement.
John: Yeah, absolutely. And someone like you, or many hospitalists kind of have looked at the Medicare guidelines and they realize that for some DRG level, whatever, it's like a three paragraph description of what goes into it. Oh, yeah, we have time to really learn all that.
Dr. Adam Robison: Doctors don't care. I tell us that you've got to optimize it and make it easy for me because I don't care if the patient needs sepsis or pneumonia or what. I know they have an infection, I know I want to treat it and I'm going to treat them appropriately. And I don't really care what came from the chart. All I care is that my patient got better because I gave appropriate care.
John: Well, this is really going to impact the CDI people, if this is going to really help them a lot. So, that's interesting. It sounds to me this business really is like a software business. And you said you're not a programmer.
Dr. Adam Robison: No.
John: So, how did you overcome that issue?
Dr. Adam Robison: Yeah. It was interesting. I had this idea, I got to fix this problem. This is a universal problem, how do I fix it? Well, oddly enough, I had been reading in our local newspaper, and I came across this guy here locally that I knew. I didn't know him at the time that he had a development for hire company. And so, I just kind of reached out to him and I said, "Hey, I got this problem. What do you think?" And he said, "Oh, yeah, I could do that. And by the way, I'm a partner of a venture fund. Let's see how this goes. And we may invest in your company if you like the idea." And so, I pitch it to them. And then that was kind of very serendipitous. We build a minimum viable product. And after they were comfortable with me and they're kind of filling me out to see if I was not as a fly by night operation, they decided to invest in the company my idea with me as well. And so, that was how we got married with a venture fund. It was very interesting how that kind of worked out. But just from a paper article, I just happened to come across it.
John: Sometimes things just work out. I always say the RAS in our brain, the reticular activating system, once you're thinking about something, it finds things that you otherwise would never notice. Now, how did you sort of protect or did you the intellectual property of the fact that you had come up with this idea? Was it built in your partnership? How do you do that?
Dr. Adam Robison: Yeah. That's an interesting thing. That's something we've still wrestled. When you develop software, you really can't patent software unless your approach is really unique. If you developed blockchain, if that's a whole new software thing, you could potentially pat that. But just patenting software, it's almost like copywriting software. You can copyright it, but anybody can write a piece of code. It's not a unique idea if you're writing this code using a typical programming language. That was a difficult problem. We have now developed a proprietary knowledge base and an ontology, if you will, that is IP and that is patentable. And we will be patenting that.
What has helped us right now and why we're five years into this right now, or almost five years into it, is what I'm doing is so ends up being very technically difficult. We have a bit of a mode around us because even the developers I hired, they didn't know how to do this because it's not a skillset you can hire off the shelf. And so, it was going through a lot of that. It's ended up been, like I said, a lot of it is been very serendipitous that turns out this isn't something you can even hire very easily for. And we had to train a lot of people up and work through it. But I think one thing we did was when protected the IP, we had that written to our contract with the development company. This is our stuff. And we had all well demark within the contract negotiations.
John: On the plus side, the more complicated it is, the less competitors you'll probably have doing it.
Dr. Adam Robison: Yeah. And we're many years into this. At this point, in this SMART on FHIR space, and unfortunately I'm saying it on a widely broadcast podcast. It is a brand new space. People are starting to understand and realize that that's how you talk to these EHRs. And what's great about it's EHR agnostic. Every EHR has to be compatible with these standards.
John: Now, one thing I'm not an expert on at all, and I've heard there's different ways of getting investors, like venture capital, angel investing, this kind of thing. But did you end up just getting the one entity to support this when you got your partner? Or did you go beyond that?
Dr. Adam Robison: No. We've had a raise. We've done a few raises now. It's been interesting. That's been a lot of work to raise money. I funded the initial development of the minimum viable product, but to actually bring on employees full time, that's not something I could afford to do. I don't have those kind of deep pockets. And so, what we did, we raised it through the initial investor network that we were introduced for that venture fund, and then raise through their network. And that got us a certain point.
And then when you raise money, every time you raise money, you're looking to say, "I'm going to raise money to hit these milestones." And so, every time we raise money, "This money was raised to hit these milestones. We've hit those milestones, now help us, we're going to raise money again to hit these milestones." You're trying to show forward momentum every time you're raising money that you've got this money. We did X, Y, Z with this money, and now we're going to do these next set of milestones, which you're going to further grow the company or hit these metrics and milestones. When we've raised money through different networks, that's how we've done it. And so, that's where we're at right now. We've raised I think about a million and a half over the last two years to expedite development and move to things forward.
John: Nice. Yeah, it's not like something you can build in your basement.
Dr. Adam Robison: No. And it's a lot of learning. I think anybody looking to start a business, just be prepared to learn a lot, because even though you may be smart as a doctor and you're very intelligent, be prepared to be very humble because you have literally no idea. You're going to have to learn a lot, fly by the seat of your pants. And that's kind of why people become entrepreneurs. They are interested in learning that stuff. But yeah, I've had to learn a lot and I've made a lot of dumb mistakes and all sorts of stuff. Lots of stories to share about that.
John: Oh, I can imagine. But it can be exciting and physicians can learn anything really as long as you have the time. That might be an issue for you as the thing gets bigger. Tell us where it is right now. You've been doing this for five years. Obviously, there was a whole ramp up. Do you have clients? Is this being live? Is it working?
Dr. Adam Robison: Oh yeah. Like I said, we have two clients right now. We have two head systems right now. The biggest thing actually, we just finished up our installation at the VA. That was a big deal to get our software working. If you're familiar with the VA or worked at VA, they have an antiquated system called CPRS. It's been around for a long time. And they've just recently built on a platform onto their CPRS that's called Lighthouse, that's compatible with SMART on FHIR. And so, our tool, we're the first third party vendor. They were looking for a solution like ours, and we met them and talked to them and they talked to some other people that could do something similar, but we were a better fit just because of our tight integration with the SMART on FHIR standards. And that has allowed us to work with them.
We just finished our deploy there. We are in the process of several other health systems. We have relationships with the biopharmaceutical company as well as a clinical research company that's looking to use our tool to help automate and augment a data retrieval from the EHR to help expedite clinical research. And so, that's what we're working now.
John: Okay. Is it to the point now where some of the medical caregivers, the physicians, APNs, whoever, are they seeing the benefits at this point?
Dr. Adam Robison: Oh yeah, we have. People love the tool. It's being used. We have some business intelligence software. We monitor the use of software, how are people using it so we can make sure we can improve it. It's not quite exponential growth in users, but we're seeing lots of people using it almost on a regular basis. And it's now becoming the way you do things now with our sites, because why would you go to a third party website or go to some other site if it can go to HER? And it's a button click, you don't have to do more work. So, it's obviously much more efficient.
John: Yeah. Nice. Do you have different formulas or different tools available over time? Or is there a list of 10 or 20 or 1,000?
Dr. Adam Robison: Yeah. Right now we have 50 or 60 medical functions automated within the chart. The ones that are commonly used. We're adding more as they come out in literature writing more. We're branching up a behavioral health space now with some of those assessments that need to be done. They're often done on paper or through PDF forms, so we're bringing them to the platform as well.
We are really just looking depending on what the clients need. Really I tell people all the time, the hardest part with this is actually just being integrated in the EHR and getting data out of it. For building a medical function, a medical calculator form, you need to get data out of HER. That's easy now. We can build that in a few days. You name it, we can build it because that's not hard. The hard part is actually integrating the EHR. That's the hard part.
John: I like that interface. That might not be the right term. But are you doing studies to see if this improves the efficiency of the physicians?
Dr. Adam Robison: Yeah, we will be. That's going to be next year. We have a couple sites that are interested in doing that. Our Richmond VA site, which we're in the process of finalizing that, they're interested in doing that kind of work. We'll be working with them to hopefully show that this is efficient, which anecdotally, of course it is. But we'll be able to show that and prove that with the data collected.
John: Yeah. The thing I hear from physicians, I have to go home and do an hour or two of notes or whatever. Now that's oftentimes the clinic, but still even hospital notes sometimes, you just can't get them done. You have to go back. Do you have a lot of capacity now? If people listen to this, we don't have a million listeners, but a lot of these physicians might still be working clinically. How do they get ahold of you or the Aimedica to learn more about this? Is it ready for that?
Dr. Adam Robison: Yeah. We're ready. We're actively taking on clients now. Like I said, we're in the process. That's going to be our big Q1 push. We had a couple clients that we're working with. Now, we're ready to go live with everybody else. And so, that's Q1, we're working on lining them up. You go to website aimedica.ai. It's easy. Easy website to remember. Just go there to contact our page, we'll be happy to do a demo for our clients.
John: Okay, aimedica.ai.
Dr. Adam Robison: That's correct.
John: All right. And if they have more questions, they can throw a note in LinkedIn to you, perhaps. Do you have time for that?
Dr. Adam Robison: Yes. Or you can email me at adam@aimedica.ai. It's very easy.
John: All right. This has been very interesting. I don't think I have a lot of other questions. Let me ask you this. What if I was going to go to my local hospital and say, "Hey, this thing sounds pretty cool." Who would I go to?
Dr. Adam Robison: You want to go to your chief information officer. They would request us through. Most of the EHR vendors have some sort of app store lineup that we would go through.
John: And what kind of presentation do you do at a system or at a hospital in terms of do you go on site, do you do it online? Do you just send some written materials? I'm just curious about how you're handling that part.
Dr. Adam Robison: Yeah. Typically, we'll do a demo on Zoom or some sort of video conferencing site. Wherever they want to use, we'll use that and we'll demo the software in a real environment so you can see how it works.
John: All right. Did I forget to ask any questions or is there anything else you want to tell us either about juggling being a hospitalist and a co-founder? Technically you do have a partner, but you were the one that created it, so I guess you are the founder. Juggling those things or about Aimedica?
Dr. Adam Robison: No. I do have hobbies that I use to not do work. I think it's important to have those so you're not doing it. I play piano. I do a couple of things to get my mind off when I have a couple minutes of downtime, which is I think important as you're looking to try to do other things with your time.
John: What do you think is going to happen with you? If this gets really big, it's going to take more and more management. Hopefully your employer at the hospital is not listening, but they probably know you got this thing going on. What do you kind of think will happen long term? Are you going to become like a CMIO in a hospital informatics in this technology? Or do both for a long time? What do you think?
Dr. Adam Robison: I'll tell you John. I actually love being a clinical doctor. I love taking care of patients. Maybe not as much as I've had in the past. I'd probably wind that down a little bit because I do work quite a bit. What I see myself probably doing the next 5, 10 years as they see guys up and it's profitable and we're able to keep lights on, is probably work for them full-time and work for my company full-time and then still see patients. Because as a chief medical officer of the company and using the tool, I find by me using the product and practicing patients, I really understand what are the problems we're trying to solve. I feel like if I remove myself from clinical practice 100%, my utility, the company to help make these tools better, I think it becomes limited as well.
John: That reminds me. Cleveland Clinic, it's a pretty big place. And I don't know about the current CEO, but the former CEO is still clinically practicing while he was running that monstrosity. So, you can keep practicing probably as long as you like, and if you're enjoying it.
Dr. Adam Robison: I do enjoy it.
John: All right. Well, I think that's all the time we have. I really appreciate that. I'm going to put the links in the show notes. Again, it's Aimedica and I've been talking to Dr. Adam Robison. I'm going to have to have you come back in a couple years and see where things have gone. And maybe by then I'll have invested in it as it'll be a listed stock or something. All right. Thanks for being here, Adam. I really appreciate it.
Dr. Adam Robison: Absolutely. Thank you.
John: Okay. Bye-bye.
Dr. Adam Robison: Bye.
Sign up to receive email reminders, news, and free stuff every week!
Enter your name and email address below and I'll send you reminders each podcast episode, notices about nonclinical jobs, information about free and paid courses, and other curated information just for you.
Transcription PNC Podcast Episode 333
How to Practice Hospital Medicine and Found a Startup
- Interview with Dr. Adam Robison
John: I'm really happy to meet today's guest because besides being a hardworking hospitalist, he's a hospital manager and leader and founder of an EHR optimization startup company, which we'll talk about for sure. Hello, Dr. Adam Robison. Thanks for coming today.
Dr. Adam Robison: Thank you so much for having me.
John: I am very happy that you're here. This is going to be very interesting. People know that I have an affinity for hospital management work. I was a CMO of a hospital and I know a lot of physicians are always trying to get away from it, but I always like to talk to people like yourself who are in hospital leadership positions and of course, also an entrepreneur. So this is going to be fantastic.
Let's just start by you telling us a little bit about your background, education and clinical career, and then we'll get into the other stuff after that.
Dr. Adam Robison: Yeah. My name is Adam Robison, like you said. I am an internal medicine doctor. I trained at the University of Louisville. I'm a big Cards fan. We did pretty good this year in football, so I was pretty happy about that. I've been working as a hospitalist clinically for the past seven and a half, almost eight years now. I work in a small community hospital out in the middle of nowhere in Idaho, in Twin Falls, Idaho. It's a great place to practice and I've been out here for a while and we are here for quite some time.
On top of that, I do work as a lead hospitalist for my group. There's about 120 providers and we cover about four or five sites right now. And so, that's been interesting work. I took that role all about almost three years ago, right at the middle kind of beginning stages of COVID. That was a very interesting time to take over as a leader of a large hospitalist group and try to navigate through that. That was good learning experience.
John: Yeah. I'm tempted to say, "Well, okay, how did you solve all those problems of people being too sick to work and not having enough PPE?" But we won't get into that. But it's been interesting, huh?
Dr. Adam Robison: Yeah, it's been interesting work. And then yeah, as you said I did found a software optimization company about five years ago, almost five years ago, 2019. And so, that's been a lot of work too. It's been a lot of interesting stuff that keeps me busy and a lot of different things to focus my attention on for sure.
John: Well, it's that second part that really got me interested. And I think our listeners are interested in things like outside work, side gig, side jobs, new careers, passive income, active income. Tell me what inspired you to start a new company? We'll see if your story jives with the others I've heard in terms of what makes people do crazy things like that.
Dr. Adam Robison: Yeah. I've been happily married for almost 20 years, and my wife, I asked her one day this crazy idea. I said, "Hey, I want to take some money. And I have this idea of how to make the EHR better." And she said, we'll have at it. And so, basically I remember sitting in front of my computer, I was using Epic at the time, using an electronic health record and going "I have to go to a third party website to finish some work. I have to go outside the EHR on a regular basis." I'm like "I have to take data from here and go kind of chart and go over here. This seems kind of stupid. Is there a better way of doing this?"
And so, initially I tried to do what I did initially within the EHR with some templates and stuff like that. And what I was looking to do wasn't actually possible within the EHR framework, the logic and the kind of advanced computing I wanted done. And so, that's what caused me to think about is there a way of doing this? And I did a lot of research. I was reading all sorts of websites, educating myself on interoperability standards. How does that work? Is there ways of doing this outside the EHR without really tight integration customized integration? And then that's what we came across. I came across something called HL7, which if you're out there is fire. And then we learned a lot about that. Really what was interesting to me was a problem that I had clinically and could I solve it. And that's what led me to found the company, realizing that we are founding a company to develop the software and to go from there. So, that's what we did.
John: That is the common thread that I've heard before. I talked to an urologist who started a company producing underwear for patients, and whatever, other different entrepreneurs. And it's always that itch that has to be scratched. You have a problem, but you're on your own, and it's like there's got to be a solution to this problem. That is very consistent. But I'm sure the way you went about it is going to be different from others because there's a myriad number of ways of doing that. We'll get into what it does exactly, but what was your next step? You had this idea, what did you do?
Dr. Adam Robison: Yeah. Whenever you have any idea, we're trying to find is there a market out there for it? Is this something that people would buy? Is this a universal or at least a broad enough problem that people run into it? I talked to lots of different physicians and people I knew, people at different hospitals I had connections with and asked them "Is this a similar problem you run into, or am I just dumb and nobody else has it?" And see if somebody else has already had a good solution for that. And realize, no, this is a fairly universal problem that everybody else has experienced. There's not a lot of solutions out there in the market. And so, that's when I go, okay. Now I don't know how to code or software. I don't have any experience with that. And I knew it would probably take me quite some time to do that. That's when I was looking around for people that knew how to do that and I was able to find some developers that had some of those skills. So, it's a lot.
There's stories about how to get licensing and getting approval through different EHR vendors. That was an interesting experience to go through, to try to convince a certain ethic that I was a company and that I wanted a license to put myself on there. Learning how to navigate all that, security protocols, yada, yada, yada. Just a lot of learning new things that I didn't know before.
John: Interesting. I just want to make sure I didn't miss another point. You checked around to see if this was a universal problem, and did that include other EHRs and the one that you were using?
Dr. Adam Robison: Yeah. I talked to a bunch of physicians because of what I have learned about, and it sounds like you talked to other doctors that have used this, but physicians are the best people at solving physician problems. Not some sort of a Silicon Valley startup by two guys out of their twenties. Like "Oh, they know how to fix our problems for us." It needs to be physicians fixing physician problems. And so, I talked to a bunch of other doctors that have been in a myriad of different EHRs. I talked to people that used the big ones in the market as well as like the VA. A bunch of colleagues worked at the VA. Is this something that you'd experienced out there? And it was a pretty universal experience.
John: Tell us about the company. Let's start there. Tell us what the company is designed to do, or what kind of product or service it provides. And then I'll come back to some of these little questions in here.
Dr. Adam Robison: Yeah. The problem that we're trying to solve was how to get data out of the EHR. It was the problem. That's a big issue. Healthcare data and how it's siloed right now is a big problem. And that's why I touched on briefly something called the HL7, which is a group, they develop these interoperability standards. It was very serendipitous timing with what I was trying to do, because during that same time, the 21st Century Cures Act had come out. And the 21st Century Cures Act for those that are familiar opens up the chart. For a lot of physicians, that meant that patients now could access certain notes in real time, which has got a lot of physicians from heartburn. All of a sudden the chart became a lot more. It used to be My Note, now it became the patient's note as well as My Note. And that was some heartburn around that.
But if you actually looked into what the 21st Century Cures Act was doing, it wasn't just making notes available to our patients, even though that's how we allowed, as physicians interpreted, it actually made the data in the EHR transparent and accessible. It had to be accessible. And it turns out, the government had adopted these things called the HL7, the standard called FHIR. And it just happened the same time while that was going on that I was trying to look to solve the same problem of how to pull data out of the EHR and do something with it meaningfully.
And so, what our tool does is it integrates directly with the patient's chart. It looks at what's going on with the patient's chart, and then pulls all the information that would be relevant to pull out, reviews the chart essentially for the user, and then provides information like MELD scores and things of that nature as well as coding recommendations all within the context of the patient's chart.
John: Interesting. Because earlier I was going to jump on the bandwagon and bad mouth some of the EHRs and EMRs. It's like you would think after helping physicians doing this for 10 or 15 years, they would've figured this out. But what you're telling me was the system wasn't necessarily ready for it and it became ready as this was implemented.
Dr. Adam Robison: Yeah. It's been interesting. I will tell a story. I have a good friend of mine who's not a physician, and him and his wife, they recently had a child and they were receiving care locally to smaller hospitals. But the child's medical condition required it to go to seek care at a tertiary care center. And what they described, even though both centers used the same EHR, they're separate EHR instances. And so, the mother and the child had two separate records that were actually very important, but then needed to talk to each other because the conditions they were seeking care for were the same at both sites. And this is just recently.
And so, for whatever reason, there hasn't been a lot of interest in the major EHR vendors to make even within their own EHR network or broadly between EHRs, that hasn't really been an interest that they've wanted to go down either maybe for economic reasons or technical reasons, for whatever reason that may be. That interoperability standard piece has been a big problem. And they haven't been willing to fix it and the Cures Act did mandate they had to open that up. But it's going to be important on companies like mine and other companies out there to do that work for them.
John: Okay. That's good. That things are progressing, and like you say, your timing was perfect. Now, I was looking at your website and reading a little bit about this, so I just want to make it clear for the listeners because some of them might actually need to use this tool at some point or tell their hospital or somebody about it. There are different kinds of metrics, there's algorithms, there's sometimes formulas that have to be used by a clinician and trying to do something. And right now you have to either do that by hand or use a separate piece of software or an app on your phone or something, and you're integrating that. So, give us more about how that works and why it's potentially a profound app.
Dr. Adam Robison: Yeah. Our two applications right now, we have two software pieces called Aicalc, Aicode that live within our Aimedica platform. And what they do is the Aicalc, you can imagine it just being your medical calculator, your cirrhosis scores, your chads bask, you name it. There's a million of these out. And they're actually propagating pretty rapidly now. With the advent of big data and smart computers, these models are coming out pretty rapidly. And our software basically looks at the patient's context, what kind of medical problems they have, is there other criteria, and says, oh, not only we'll calculate, oh, patient's cirrhosis, you'll need a MELD score. And then we'll give you the most up to date MELD score, give you that sort of stuff. You don't have to click around. You just click the button. You don't have to think of why you need the MELD score. We'll just review the patient chart and give you all the relevant scores and pull the data in to calculate the scores for you automatically.
Again, one thing I say to people all the time, I use this tool every day in my practice. I developed it for myself, and so I'm constantly trying to improve it. And then we also do the same thing with medical diagnosis codes. We try to optimize the best highest weighted codes for DRG waiting for hospitalizations, which is important because we can look at a patient's chart and say, "Oh, the patient actually meets sepsis criteria and you are treating of infections. That should be the code you put in the chart. Let's provide that." And that has a huge ramification. And anybody knows in the healthcare space, those DRG waiting is huge for hospitalizations reimbursement.
John: Yeah, absolutely. And someone like you, or many hospitalists kind of have looked at the Medicare guidelines and they realize that for some DRG level, whatever, it's like a three paragraph description of what goes into it. Oh, yeah, we have time to really learn all that.
Dr. Adam Robison: Doctors don't care. I tell us that you've got to optimize it and make it easy for me because I don't care if the patient needs sepsis or pneumonia or what. I know they have an infection, I know I want to treat it and I'm going to treat them appropriately. And I don't really care what came from the chart. All I care is that my patient got better because I gave appropriate care.
John: Well, this is really going to impact the CDI people, if this is going to really help them a lot. So, that's interesting. It sounds to me this business really is like a software business. And you said you're not a programmer.
Dr. Adam Robison: No.
John: So, how did you overcome that issue?
Dr. Adam Robison: Yeah. It was interesting. I had this idea, I got to fix this problem. This is a universal problem, how do I fix it? Well, oddly enough, I had been reading in our local newspaper, and I came across this guy here locally that I knew. I didn't know him at the time that he had a development for hire company. And so, I just kind of reached out to him and I said, "Hey, I got this problem. What do you think?" And he said, "Oh, yeah, I could do that. And by the way, I'm a partner of a venture fund. Let's see how this goes. And we may invest in your company if you like the idea." And so, I pitch it to them. And then that was kind of very serendipitous. We build a minimum viable product. And after they were comfortable with me and they're kind of filling me out to see if I was not as a fly by night operation, they decided to invest in the company my idea with me as well. And so, that was how we got married with a venture fund. It was very interesting how that kind of worked out. But just from a paper article, I just happened to come across it.
John: Sometimes things just work out. I always say the RAS in our brain, the reticular activating system, once you're thinking about something, it finds things that you otherwise would never notice. Now, how did you sort of protect or did you the intellectual property of the fact that you had come up with this idea? Was it built in your partnership? How do you do that?
Dr. Adam Robison: Yeah. That's an interesting thing. That's something we've still wrestled. When you develop software, you really can't patent software unless your approach is really unique. If you developed blockchain, if that's a whole new software thing, you could potentially pat that. But just patenting software, it's almost like copywriting software. You can copyright it, but anybody can write a piece of code. It's not a unique idea if you're writing this code using a typical programming language. That was a difficult problem. We have now developed a proprietary knowledge base and an ontology, if you will, that is IP and that is patentable. And we will be patenting that.
What has helped us right now and why we're five years into this right now, or almost five years into it, is what I'm doing is so ends up being very technically difficult. We have a bit of a mode around us because even the developers I hired, they didn't know how to do this because it's not a skillset you can hire off the shelf. And so, it was going through a lot of that. It's ended up been, like I said, a lot of it is been very serendipitous that turns out this isn't something you can even hire very easily for. And we had to train a lot of people up and work through it. But I think one thing we did was when protected the IP, we had that written to our contract with the development company. This is our stuff. And we had all well demark within the contract negotiations.
John: On the plus side, the more complicated it is, the less competitors you'll probably have doing it.
Dr. Adam Robison: Yeah. And we're many years into this. At this point, in this SMART on FHIR space, and unfortunately I'm saying it on a widely broadcast podcast. It is a brand new space. People are starting to understand and realize that that's how you talk to these EHRs. And what's great about it's EHR agnostic. Every EHR has to be compatible with these standards.
John: Now, one thing I'm not an expert on at all, and I've heard there's different ways of getting investors, like venture capital, angel investing, this kind of thing. But did you end up just getting the one entity to support this when you got your partner? Or did you go beyond that?
Dr. Adam Robison: No. We've had a raise. We've done a few raises now. It's been interesting. That's been a lot of work to raise money. I funded the initial development of the minimum viable product, but to actually bring on employees full time, that's not something I could afford to do. I don't have those kind of deep pockets. And so, what we did, we raised it through the initial investor network that we were introduced for that venture fund, and then raise through their network. And that got us a certain point.
And then when you raise money, every time you raise money, you're looking to say, "I'm going to raise money to hit these milestones." And so, every time we raise money, "This money was raised to hit these milestones. We've hit those milestones, now help us, we're going to raise money again to hit these milestones." You're trying to show forward momentum every time you're raising money that you've got this money. We did X, Y, Z with this money, and now we're going to do these next set of milestones, which you're going to further grow the company or hit these metrics and milestones. When we've raised money through different networks, that's how we've done it. And so, that's where we're at right now. We've raised I think about a million and a half over the last two years to expedite development and move to things forward.
John: Nice. Yeah, it's not like something you can build in your basement.
Dr. Adam Robison: No. And it's a lot of learning. I think anybody looking to start a business, just be prepared to learn a lot, because even though you may be smart as a doctor and you're very intelligent, be prepared to be very humble because you have literally no idea. You're going to have to learn a lot, fly by the seat of your pants. And that's kind of why people become entrepreneurs. They are interested in learning that stuff. But yeah, I've had to learn a lot and I've made a lot of dumb mistakes and all sorts of stuff. Lots of stories to share about that.
John: Oh, I can imagine. But it can be exciting and physicians can learn anything really as long as you have the time. That might be an issue for you as the thing gets bigger. Tell us where it is right now. You've been doing this for five years. Obviously, there was a whole ramp up. Do you have clients? Is this being live? Is it working?
Dr. Adam Robison: Oh yeah. Like I said, we have two clients right now. We have two head systems right now. The biggest thing actually, we just finished up our installation at the VA. That was a big deal to get our software working. If you're familiar with the VA or worked at VA, they have an antiquated system called CPRS. It's been around for a long time. And they've just recently built on a platform onto their CPRS that's called Lighthouse, that's compatible with SMART on FHIR. And so, our tool, we're the first third party vendor. They were looking for a solution like ours, and we met them and talked to them and they talked to some other people that could do something similar, but we were a better fit just because of our tight integration with the SMART on FHIR standards. And that has allowed us to work with them.
We just finished our deploy there. We are in the process of several other health systems. We have relationships with the biopharmaceutical company as well as a clinical research company that's looking to use our tool to help automate and augment a data retrieval from the EHR to help expedite clinical research. And so, that's what we're working now.
John: Okay. Is it to the point now where some of the medical caregivers, the physicians, APNs, whoever, are they seeing the benefits at this point?
Dr. Adam Robison: Oh yeah, we have. People love the tool. It's being used. We have some business intelligence software. We monitor the use of software, how are people using it so we can make sure we can improve it. It's not quite exponential growth in users, but we're seeing lots of people using it almost on a regular basis. And it's now becoming the way you do things now with our sites, because why would you go to a third party website or go to some other site if it can go to HER? And it's a button click, you don't have to do more work. So, it's obviously much more efficient.
John: Yeah. Nice. Do you have different formulas or different tools available over time? Or is there a list of 10 or 20 or 1,000?
Dr. Adam Robison: Yeah. Right now we have 50 or 60 medical functions automated within the chart. The ones that are commonly used. We're adding more as they come out in literature writing more. We're branching up a behavioral health space now with some of those assessments that need to be done. They're often done on paper or through PDF forms, so we're bringing them to the platform as well.
We are really just looking depending on what the clients need. Really I tell people all the time, the hardest part with this is actually just being integrated in the EHR and getting data out of it. For building a medical function, a medical calculator form, you need to get data out of HER. That's easy now. We can build that in a few days. You name it, we can build it because that's not hard. The hard part is actually integrating the EHR. That's the hard part.
John: I like that interface. That might not be the right term. But are you doing studies to see if this improves the efficiency of the physicians?
Dr. Adam Robison: Yeah, we will be. That's going to be next year. We have a couple sites that are interested in doing that. Our Richmond VA site, which we're in the process of finalizing that, they're interested in doing that kind of work. We'll be working with them to hopefully show that this is efficient, which anecdotally, of course it is. But we'll be able to show that and prove that with the data collected.
John: Yeah. The thing I hear from physicians, I have to go home and do an hour or two of notes or whatever. Now that's oftentimes the clinic, but still even hospital notes sometimes, you just can't get them done. You have to go back. Do you have a lot of capacity now? If people listen to this, we don't have a million listeners, but a lot of these physicians might still be working clinically. How do they get ahold of you or the Aimedica to learn more about this? Is it ready for that?
Dr. Adam Robison: Yeah. We're ready. We're actively taking on clients now. Like I said, we're in the process. That's going to be our big Q1 push. We had a couple clients that we're working with. Now, we're ready to go live with everybody else. And so, that's Q1, we're working on lining them up. You go to website aimedica.ai. It's easy. Easy website to remember. Just go there to contact our page, we'll be happy to do a demo for our clients.
John: Okay, aimedica.ai.
Dr. Adam Robison: That's correct.
John: All right. And if they have more questions, they can throw a note in LinkedIn to you, perhaps. Do you have time for that?
Dr. Adam Robison: Yes. Or you can email me at adam@aimedica.ai. It's very easy.
John: All right. This has been very interesting. I don't think I have a lot of other questions. Let me ask you this. What if I was going to go to my local hospital and say, "Hey, this thing sounds pretty cool." Who would I go to?
Dr. Adam Robison: You want to go to your chief information officer. They would request us through. Most of the EHR vendors have some sort of app store lineup that we would go through.
John: And what kind of presentation do you do at a system or at a hospital in terms of do you go on site, do you do it online? Do you just send some written materials? I'm just curious about how you're handling that part.
Dr. Adam Robison: Yeah. Typically, we'll do a demo on Zoom or some sort of video conferencing site. Wherever they want to use, we'll use that and we'll demo the software in a real environment so you can see how it works.
John: All right. Did I forget to ask any questions or is there anything else you want to tell us either about juggling being a hospitalist and a co-founder? Technically you do have a partner, but you were the one that created it, so I guess you are the founder. Juggling those things or about Aimedica?
Dr. Adam Robison: No. I do have hobbies that I use to not do work. I think it's important to have those so you're not doing it. I play piano. I do a couple of things to get my mind off when I have a couple minutes of downtime, which is I think important as you're looking to try to do other things with your time.
John: What do you think is going to happen with you? If this gets really big, it's going to take more and more management. Hopefully your employer at the hospital is not listening, but they probably know you got this thing going on. What do you kind of think will happen long term? Are you going to become like a CMIO in a hospital informatics in this technology? Or do both for a long time? What do you think?
Dr. Adam Robison: I'll tell you John. I actually love being a clinical doctor. I love taking care of patients. Maybe not as much as I've had in the past. I'd probably wind that down a little bit because I do work quite a bit. What I see myself probably doing the next 5, 10 years as they see guys up and it's profitable and we're able to keep lights on, is probably work for them full-time and work for my company full-time and then still see patients. Because as a chief medical officer of the company and using the tool, I find by me using the product and practicing patients, I really understand what are the problems we're trying to solve. I feel like if I remove myself from clinical practice 100%, my utility, the company to help make these tools better, I think it becomes limited as well.
John: That reminds me. Cleveland Clinic, it's a pretty big place. And I don't know about the current CEO, but the former CEO is still clinically practicing while he was running that monstrosity. So, you can keep practicing probably as long as you like, and if you're enjoying it.
Dr. Adam Robison: I do enjoy it.
John: All right. Well, I think that's all the time we have. I really appreciate that. I'm going to put the links in the show notes. Again, it's Aimedica and I've been talking to Dr. Adam Robison. I'm going to have to have you come back in a couple years and see where things have gone. And maybe by then I'll have invested in it as it'll be a listed stock or something. All right. Thanks for being here, Adam. I really appreciate it.
Dr. Adam Robison: Absolutely. Thank you.
John: Okay. Bye-bye.
Dr. Adam Robison: Bye.
Sign up to receive email reminders, news, and free stuff every week!
Enter your name and email address below and I'll send you reminders each podcast episode, notices about nonclinical jobs, information about free and paid courses, and other curated information just for you.
Disclaimers:
Many of the links that I refer you to are affiliate links. That means that I receive a payment from the seller if you purchase the affiliate item using my link. Doing so has no effect on the price you are charged. I only promote products and services that I believe are of high quality and will be useful to you. As an Amazon Associate, I earn from qualifying purchases.
The opinions expressed here are mine and my guest’s. While the information provided on the podcast is true and accurate to the best of my knowledge, there is no express or implied guarantee that using the methods discussed here will lead to success in your career, life, or business.
The information presented on this blog and related podcast is for entertainment and/or informational purposes only. I do not provide medical, legal, tax, or emotional advice. If you take action on the information provided on the blog or podcast, it is at your own risk. Always consult an attorney, accountant, career counselor, or other professional before making any major decisions about your career.
Leave A Comment
You must be logged in to post a comment.