Designing for Health: Interview with Farhan Ahmad and Jon Keevil, MD [Podcast]

Emerging technologies have vast, expansive abilities within the healthcare landscape. However, a new asset can be needlessly complex and too overwhelming to gain wide adoption amongst clinicians. In order for wholesale implementation to take place, new software or technologies need to balance increased capabilities with the simplicity and accessibility for use in day-to-day clinical operations. The right mix of technology and simplicity is essential to ensure a better clinical experience for caregivers and patients alike.

On today’s episode of In Network's Designing for Health podcast feature, Nordic Chief Medical Officer Craig Joseph, MD, talks with Farhan Ahmad and Jon Keevil, MD, co-founders of HealthDecision, a company that provided clinical diagnostic decision-making tools. Their discussion centers around how they balanced Farhan’s software expertise with Jon’s clinical experience to create a powerful analytical asset, while also being accessible and simply designed for healthcare providers. They also talk about the concept of sludge versus nudge, the challenge of interpreting medical guidelines, and how using data to inform patients can enhance clinical decision making and outcomes.

Listen here:



In Network's Designing for Health podcast feature is available on all major podcasting platforms, including Apple PodcastsAmazon MusiciHeartPandoraSpotify, and more. Search for 'In Network' and subscribe for updates on future episodes. Like what you hear? Make sure to leave a 5-star rating and write a review to help others find the podcast.

Want to hear more from Dr. Joseph? Order a copy of his book, Designing for Health.


Show Notes:

[00:00] Intros

[07:32] The beginnings of HealthDecision

[14:32] Inside the cardiovascular risk score

[19:46] Combining software with clinical knowledge

[21:38] Sludge vs nudge

[27:05] Designing for different types of audiences

[29:07] Interpreting medical guidelines

[36:00] What Jon and Farhan are doing now

[38:46] Well-designed things that bring Jon and Farhan joy

[44:04] Outros



Dr. Craig Joseph: Welcome Jon and Farhan to the podcast. I'm so excited to have you both here. We're doing something we've never done before. You two are in the same room at Nordic’s International Galactic Headquarters, and I am in my house, which is far away. And so I don't want, I want to make sure that you don't say anything bad about me that we're not catching on tape. Does that sound fair?

Farhan Ahmad: Sure. Yeah, yeah.

Dr. Craig Joseph: All right. So, Farhan, why don't we kind of get started with you? You and I met a long time ago when we both worked for an unnamed electronic health record software vendor based in Verona, Wisconsin. And, you were, doing kind of EDI, kind of software development or electronic data exchange. Can you kind of just, tell us how you got there, what you do before and what you did a little bit afterwards, and then we'll get to Jon and see how you two overlap.

Farhan Ahmad: Sure. Yeah so, my background is all on the computer science side. I was professionally trained to have a bachelor's in computer science, went down the software development path and then, from work for a biller, a billing software company. And then from there, I applied to the said vendor earlier, and I applied as a software developer, but then, they were looking for-they saw that I had a little bit of people side to me, so they instead put me in the data interchange side, which does require talking to people more than the software development side does. So I did the data exchange. So, making the EHR talk to many other systems within the healthcare organization. I did that for a few years and then moved on to consulting on technology development. Worked with two other founders on a consulting company that focused on software development, implementation side of things. So, I got a pretty unique perspective from both the management side, but then also, actually putting the software to use, from the user's perspective, too. My focus there was really more on the development side of things. So really taking, the package and combining those things and taking them to the next level that, initially maybe they weren't fully meant to do, but getting the maximum out of it. So that company got acquired after we were around more than 130 consultants. From there on, during that company, I ran into Jon and, that's where we kind of started crossing paths. And then after I, went through the acquisition integration, that's where, I ended up with HealthDecision, and we joined together.

Dr. Craig Joseph: That is awesome. And I'm excited to hear that you were not allowed to do development as a developer, and that is, we when we were preparing for this, I've heard that story very often, from people who kind of went in, just decided that they were going to be writing code and, somehow, some test or some interaction, said, oh, this is a developer who can develop and they can talk to people who are not developers, which is a little complicated. And, and I think, to some extent that helped you in, design further on, but we'll put a pause here and now I think, Jon, why don't you tell us your story? I think it's very, similar and dissimilar at the same time.

Dr. Jon Keevil: Yeah, there's a little overlap. There's, my background, you know, undergraduate was computer science and engineering. I wasn't sure what I wanted to do. After three years, I took some time off. I worked in a thick film chip resistor company, that made little resistors and did some music, and then decided that I really wanted to be more people focused and went into medicine and, finished medical school and came to Wisconsin and did internal medicine and cardiology training, and so became, you know, went through my cardiology process of education and then started as an attending in in the year 2000. And there's actually a modest, interesting sort of overlap in cardiology and engineering because it's a lot of pumps and electricity, and there's a lot of people that think in those same ways, and, you know, when I came into the Department of Medicine, I knew what an operating system was. So, I was the computer expert. You know, it's like I was clearly, you know, got on the committees that were dealing with EHRs and such, and in the process of learning about cardiology, it became obvious there were really complex questions that were getting more complex. I think we've discussed this together, where the Framingham study, which started in the late 40s and has been continuing, is really the original big data study, I think, in medicine, where it's gathered real insight and sort of holds within it some level of intelligence. And you can calculate a predicted risk of cardiovascular disease. And that became a required step for clinicians. We actually had to do the math, do the calculation, and then guide the patient. And whether they get statins or not based on that calculation. And my brain was always thinking this is too complicated. This is, you know, even as a specialist, I could do it, but there's lots of steps. And then my wife and my sister, both family doctors, and it's really hard for them. And so, the idea of having a tool that could help that process, help, you know, figure out these steps, maybe even be connected to the EHR, became a dream back in early 2000, and over the next ten years or so, I developed a tool that worked locally at UW. And then because I was on those committees, I actually was part of the group that got interested in Farhan’s company and, brought him into UW to do some custom software. We met through that interaction. And then later as my project got more and more involved and I needed other people, he started supplying me with programmers and people. And then when I decided to leave medicine and go into this full time, he joined me as sort of second in command and, and was, you know, somebody who really knew how to program. I just kind of knew about programing at that point. but that was our that was our connection.

Dr. Craig Joseph: That's fair. And so that when you started your company was it HeartDecision?

Dr. Jon Keevil: Yeah. Initially I called it HeartDecision, as in 2004, it basically had a separate checking account. And that was to allow me to hire some people to do the initial real programing that took my spreadsheet and put it on the web. And then that also allowed us to do our first integration. This was back at a time when the doors were a little more open about people trying to do custom connections, and they actually gave us some guidance about how our hospital could, you know, share some data back and forth and do that. So, I was working with the IT people at UW and the EHR to make those early connections, to share the data into the app and then and then use the app within the EHR.

Dr. Craig Joseph: Okay. So ultimately the company became it was called HealthDecision. And so, you were expanding a little bit outside of cardiology. tell us a little bit, though, how you started. So, once it sounds like you kind of, started to take off when you brought in a professional developer, and someone else who, you know, had lots of experience and, kind of making sure software is usable and can be implemented and operationalize. How did that whole process kind of take off?

Dr. Jon Keevil: You know, it was amazing. I mean, looking back on it, it was such a rich and lucky kind of combination of things because as a physician working inside the system, I then have access to a quality grant that our IT department put out, and they had a small stipend to praise for some programing, but it also came with developer time internally. And so, I submitted a proposal that I could take the Framingham score and make it operational and make it useful for patients and doctors. And they said, cool, we like that. And so, everybody was kind of working together. And when you call the EHR company and you're a doctor working in that hospital, they're like, oh, you're a customer, you're yourself. Come on in. We can talk to you about it and help you. And then Farhan had this wonderful bridge of knowledge where he had worked with the EHR, but he also knew how to work with hospitals. And so, he was kind of the technical link. And then I had all this access to all these other colleagues that were starting to use the tool. So, we'd build the tool and then we'd go and sit with colleagues that were using it or had used it recently and get feedback, and then we'd build some more and then we'd go back and get feedback. And I was under no rush because I was just being a doctor. And this was just being helpful. And so, we actually had quite a few years of just sort of gentle development where we kept kind of improving it. And, and we weren't really thinking about the, the commercial side of our own company at that point. And so, it was this real sort of luxury of development of, you know, getting a lot of people to kind of be curious about it and give some feedback. And then, you know, a week later we'd have an improvement and show it to them and say, hey, you, you suggested this, and it's, now it's better. Well, that was a lot of fun. It was kind of a cool time.

Dr. Craig Joseph: So Farhan, from your perspective, did you did you have other developers or was this just, it was a pretty much a one person show.

Farhan Ahmad: Oh, going back to your comment about developers not being able to develop, one thing I was just going to add is, knowing how to write code is a blessing and a curse because and especially when you, trained professionally to be a developer and then later on, you're not able to do it right, and it becomes hard to keep your hands away from the keyboard and not to do it, but really to, scale, you do have to start doing that. so that was one thing was like, yes, we did have developers, initially under the other company that we had, Jon had hired one of my developers, that was focusing on, the HealthDecision product at that point. And then later on, when we joined forces, at that point, we hired more people, to help code and do most of this. I was still dabbling very little, but most of the work was done by the, the really good programmers that we had hired at that time. So, it might look like all of a sudden, like when I joined the company, that, things started moving faster. But really, it was at that point that Jon decided to focus on it full time himself, too, that, we really started to pick up speed and be able to focus. That gave us the, the speed and boost that we needed.

Dr. Craig Joseph: So, when you, when you started, Farhan, with Jon, even then, you were really not allowed to be the person getting in there writing the code. And that's because there was so much of the, you know, understanding how IT shops worked at hospitals and, you know, kind of being playing nicely in the healthcare sphere that sometimes we as physicians, you know, especially when you're practicing full time, Jon, like you don't you don't see a lot of the behind the scenes stuff that, that Farhan saw both as, as an Epic employee, there I said it, and as a founder and owner of a consulting company in the healthcare IT space. Was that disappointing, or you were already well on your way of kind of being in management at that point.

Farhan Ahmad: I had that understanding, and it's like, one of those things where, like the curse part of it, you know, that's what you need to do. That's what you need to not be doing, is doing the programing part of it. And one of the other things to note is this is, back when the EHRs did not have the app stores that they do now. And, and over time that started coming. So, it presented an even more challenge, to figure out how to do the data integration, which was one of the core value propositions that we had for the product. And then we were also working through some platform upgrades at, early on and getting the company together. So, there was plenty of other stuff to be doing anyway.

Dr. Jon Keevil: For me, it felt like a two for one to actually, because those early stages, I was really just a customer of Farhan’s company because I had hired, I had gone to him and said, hey, I've got this project, can you work on it? And he said, sure. And I think for the first few weeks you tried to be the programmer on it. And then it became obvious this was going to be a lot of work. And so you hired somebody who then became the, he was an ended up being a superstar and a long term employee of ours. But he started off as a fairly new programmer with Farhan. And so I got him full time. But he would always go to Farhan for guidance and help. And so I was getting like, Farhan's fingerprints still on his growth and his, what he was learning. And so that seemed like a pretty good arrangement from my perspective.

Dr. Craig Joseph: That's a good deal, so, let's talk about the software that you all created. So it started off with essentially just, hey, we're going to take all the discrete parts that contribute to this, this, cardiovascular risk score. And we're going to automatically calculate that for physicians. And that's just a number. Right. And physicians interpret. Can you give us a little bit more about that score and how physicians, what physicians do with it, if anything?

Dr. Jon Keevil: Yeah, sure. No, the whole thing started probably in about 2001. The guidelines that came out in 2001 for cholesterol management recommended, and, you know, expected, that clinicians were going to do this scoring process. And then they had thresholds of sort of what different levels of cholesterol and what level of score matched up to be in the category of, oh yeah, you probably should start a statin medication. And so, it became this, I actually wrote a paper about this in the Journal of Circulation. I did research on the country's, sort of population databases that, that sort of showed this big like six by five grid of the, you know, six different score levels and five different cholesterol levels. How that all kind of matches out and what number of the population was in each one of these cells in the grid. And so, it's a little bit of a complicated table. So it's not only the score, but a score of say, 12% means one thing for one person if they're otherwise at low risk. And, and if their cholesterol is really high or really low and it may mean something different for someone else. So we had to combine those two kind of assets. What's your cholesterol level that partly goes into the score? But it also then triggers the treatment recommendations. And so we tried to build the software to not only show the score, but then show what the guidelines would say. So make it clear which category this person was in. And a lot of clinicians knew that. And we're learning that and figuring that out. So that was familiar to them. And then we added this whole other layer with all the scientific data that existed on statins, is we could then predict how much their risk would drop with a statin. So we actually took all the statins that are out there, and you could predict what the LDL percent drop was. And then through all the literature, we actually have a pretty good idea of predicting what the risk does. And so then you could teach a clinician and the patient how much their risk would drop if they started the statin and how to kind of put that in context. And we actually built a whole graphical way of communicating this with little groups of, you know, little icon arrays, so-called, with, you know, 100 people or a thousand people, depending on the risk. And if you took a statin or did some other action like do mammograms or something for breast cancer, we built one of those later to or start an anticoagulant for atrial fibrillation. We built this this sort of visual model where we could show, here's your risk, meaning here's the chance out of a population of 100 people like you, that such, you know, some sort of bad thing might happen. And if you do this treatment or this therapy or take the screening test, here's what we expect. That risk will drop, how it will drop. And here's some side effects that might show up. Or here's some bad things that could happen. And so we ended up sort of demonstrating visually the yin and the yang of these decisions, which ended up being really helpful for people, both clinicians who often felt like, hey, I'm being a better clinician because I'm being more thoughtful in the way I describe this. And they become better educators because they can kind of show the patient their specific situation. And then the patients become more informed consumers when they sort of make a choice to either do or not do, treatment.

Dr. Craig Joseph: So what was your vision when you first started? Was it just to kind of calculate a score and be done with it? But then you saw that you needed to, I'm curious how you kind of moved into the, the graphical part of it. And because neither of you are, I believe, have a graphical arts, background.

Dr. Jon Keevil: No, but I have a teaching background, you know, so I was the director of the second year medical school course for 13 years in my career and, and a very and a lot of, you know, a lot of primary care, but a lot of preventive cardiology is educating patients. And if you have an idea of something they might want to do and they don't understand your reasoning for that, there's a lot of discussion that goes into just kind of teaching them of why you think what you think and why you think a statin might be useful or not useful. And I would often have primary care doctors come up to me and say, your tool is so helpful, I could easily show this patient they didn't need a statin and they'd show a hundred people. And your risk is 3%. And if we take a statin over ten years, it'll drop to 2%. And look, one person got benefit and the other 99 didn't. And that was so helpful for people to kind of, you know, they have these light bulbs go on. I was like, oh, okay, I get it. I get how important this is or not important this is.

Dr. Craig Joseph: Yeah. Well, I you know, that's almost a 50% reduction from 3 to 2.

Dr. Jon Keevil: You could even say it's 33%.

Dr. Craig Joseph: 33%, reduction. Exactly. Yet that's probably not as significant. If it's three people versus two people. So, I think that was kind of the yeah, that was the brilliant aspect. Farhan, were there times where you wanted to kind of apply your IT skills, your technical skills? and they kind of ran into the, the clinical needs. In other words, you know, Jon said, oh, I just got this feedback, and this is great. We got to do it this way. You know, like, we could, but then there's going to be lots of other problems. Any good stories like that? And I'm looking specifically for fistfights that might have broken out right between you two.

Dr. Craig Joseph: Yeah.

Farhan Ahmad: Yeah, absolutely. Early on it was like that. And that was mainly learning for myself because, when you're doing software development, it's very black and white. And you put in rules to do something a certain way and, very quickly, one of the things I learned, and then this is like even when I was at the EHR company and some of the other cases, it's that, the users in this case on like one half of the users they’re actual physicians many times they're multiple PhD level of folks that we were talking to.

Farhan Ahmad: So I don't need to be holding their hand and saying, no, you must do this. If they're not doing something very likely, there's a good reason behind it. So, and this is one thing where, just going back to like, your book that you've written, there's a tendency for us as software developers to create more of the sludges or really forced rules rather than the nudges that the users needed and that's one of the things where, I was really lucky to have partnered with Jon because he brought that perspective from, his background and the expertise that he had to say that, no, we don't need to tell that to the physician because they already know it. And really, it's the guidance. And if they're going in a certain direction, there's very likely a good reason so we need to give them the nudge, but not force them one way or not or another.

Dr. Craig Joseph: So thank you for bringing out the sludge versus nudge. And let me be clear that that is certainly not my brilliance. But, how did you kind of, Jon, how do you determine the difference between the sludge and the nudge? And let's maybe give a little, discussion about or a little explanation about what is what is a nudge, at least in your mind, for either of you. And what is sludge? And how do you know when you see it?

Dr. Jon Keevil: I'm not familiar with the sludge term yet. I read a whole book called nudge, but ...

Farhan Ahmad: Well, we have the author right here.

Dr. Craig Joseph: So, so then I'll go. Sludge is actually part of that, of that book, about nudges and, small part. But, you know, I think, most of us would talk when we talk about sludge. It's mandatory kinds of clicks, you know, showing you something that you, you don't need, but there's no way for you to get around it. It's kind of like, think about clinical decision support tools that pop up to remind you to do something that you were just about to do. And it just kind of, you know, gets in the way and it doesn't actually help.

Dr. Jon Keevil: Yeah. No, I, I think where a lot of this comes from, you know, my perspective was partly informed that my wife is a family doctor. We went through medical school together. We got about the same score on, on boards and, you know, examinations and that sort of thing. I think our area of under the curve of intelligence is probably pretty similar. And then she went into family practice and knows a little bit about like everything, you know, she delivered babies. She did small surgeries. She did all this stuff. I did cardiology and I really drilled down into one little specific area. And we actually recognize, you know, sidelight on our, our relationship. We kind of recognize at one point that often the cases we talked about were cardiac cases, because that would be something both of us knew something about. But I was always the expert in that conversation. And I think as a specialist, you can get into that, that the blinders get on where you don't recognize that the person who you're giving a recommendation to has this much grander vision of what the patient's up to. And so I think I really respected as a specialist in a large health system, that I just was seeing one, one narrow area of the patient's kind of needs. And they came to me about this cardiac risk, but maybe they had other drugs that would react with statins, or they had some other thing in their family or something else that was going on that was just a bigger deal. So I think I gained a lot of respect for when people choose not to do whatever it was that I recommended. And I sort of had this general adage of if a patient, if I've had the opportunity to really tell a patient why I think something and they seem to get it and understand it, and then they still decide not to do it, I'm fine with that. That's totally okay. So I came with that sort of, you know, orientation. And I also came to this orientation that my wife's a smart physician. And so she's choosing not to do something. There's probably a good reason. And so I often felt I was kind of defending the clinicians, and I didn't have to work very hard to defend them to Farhan, but to people like Farhan, you know, Farhan’s position. You know, I've been on committees where people are designing things and they're designing something for the doctor. And I realized, like, you know, you have to change your mindset. The mindset of that physician is he's trying to make sure the patient stays safe. You know, he or she is really working to protect the patient. And so that, you know, they don't die because the electronic record did the wrong thing or something and that kind of opened up the eyes of some programmers and like, oh, I didn't think of it that way. Right? Yeah, that's a good point, I guess. So, so I often felt like I was speaking for that side of it. It's like the people who are working really hard late nights, you know, always. And just, you know, a sludge, another piece of sludge is just going to put you over the edge. If it uses up the last three cycles of brain energy you had for the day, and you're kind of toasted by that sort of stuff. So, I, I think one of our missions really became get rid of that stuff and make it as clean and simple and as efficient as we possibly can, and show the information if people want it. So we tried to have a really clean screen, but then have a lot of little things. We called info buttons, where if you wanted to learn more about it, click here and we'll tell you more. But if you if you see the recommendation and it makes perfect sense to you and you're just moving on, then let's not bother you with explaining what level of recommendation it is or why. You know you'll be criticized if you don't do it, or some other sort of layer that may not apply to this personal right.

Farhan Ahmad: And one of the other, things that is, is unique when you start integrating with third party systems is what happens when the data is not there. And I know we spent plenty of time thinking through that, trying to figure it out. Do we assume the data does not exist? Do we assume that it was that, for instance, a test was never done, or it was negative or if, or positive, depending on what situation you're looking at? And some of those could end up very easily into nudges that we tried our best to keep them as, as nudges, not sludges.

Dr. Craig Joseph: Yeah. That that makes that makes a lot of sense. I find that, you know, sometimes it's black and white, but usually so often it's gray. And who's your intended audience, right. So it a cardiologist? Is it a family doc? Is it a nurse practitioner? Is it a nurse? Is it the patient? And so you had to develop for all of those different audiences. Did you run into kind of technical problems? Obviously the first, the EHR that you were kind of integrating with was Epic. Were there other EHRs at the time that you were integrating with and, any insights there about how different, vendors had different philosophies?

Farhan Ahmad: Yeah. I think before we go there, I want to just, touch on one thing and maybe Jon can expand on this part of it. One of the things, because we were dealing with guidelines, some things that we ran into as many times the guidelines did not cover everything. And at the same level of detail that we had to then go in and interpret and say, okay, you're not looking at these different races or these different ethnicities. And that presented a pretty unique challenge. Maybe, Jon, if you want to expand on that a bit.

Dr. Jon Keevil: It was interesting. We often had to kind of reverse engineer the guidelines, because I wanted to be able to take the recommendations that a, you know, a guideline said for each kind of combination of person. And I was hoping that every patient that you had data in, you would get some sort of feedback. And occasionally we kind of came across, you know, holes in the data where, you know, most likely the guideline just never really had enough population of this particular combination of this age of male, you know, in this class of whatever and we just they didn't have any information about it. And so, it was really intriguing because it gave us some more insight into what it takes to write a guideline and all the, you know, those layers that go into it. But that was you know, you had to sort of pick it apart. There's a quote I like that you have to really understand complexity before you can make it simple. And so we had to really dig into all the layers of a guideline before we could then very simply present the one piece of the guideline that this patient needs.

Dr. Craig Joseph: Did you have any problems as you were interpreting these guidelines from the Journal article? and actually technically making that happen, in terms of that implementation?

Farhan Ahmad: Yeah. And that's one where I was glad to have Jon by my side, because he's really good at it, because when you look at all the details, some of these guidelines are wrote like 20 plus pages with a lot of tables in it. So it's not like you want to be guiding the physician through each and every step of that, and instead you want them to come to the final conclusion so that's where they go, which is just as a side note, that's another thing is like, that, as a technologist, my desire is to kind of show exactly what's happening, but that from the physician perspective, they already understand it. So there's no point in showing how we're going about it unless you do want to dig into it. But a majority of times you just want to see what the outcome ends up being.

Dr. Craig Joseph: Farhan that story, really actually hits home with me. When I was working at Epic, one of the things that I was tasked with doing is trying to build in the pediatric immunization guidelines into the electronic health record, which sounds pretty straightforward until you actually read the guidelines and, we ran into things and I had developers come to me and say, well, what does this mean, exactly? Tell me exactly what this means. And sometimes, something would say, well, you have to wait 30 days between doses if you wait 29, if you give the second dose at 29 days, it doesn't count. Got to wait 30 days. But in the very next paragraph, it would say one month. And then, you know, developers are like well is it is it 30 days and is it one month? But if it's one month, you know, what happens when it was March 31st and now it's April, what's one month exactly? Do I just go with the same day? Those are very valid questions and computers as you have noted, are ones and zeros. And so there's really no gray that's allowed there. So yeah, that that is something that I think is getting better in the literature. Jon, would you say that have you seen improvements in terms of guidelines being written with the idea of being able to actually operationalize them in technology?

Dr. Jon Keevil: I think so, yeah. Because I think a lot of guidelines then got translated into quality metrics. And anybody who was building a quality metric, whether they were on the computer side or not, kind of recognized, like, wait, do I count this as a win or not, you know, is this good or not? And then they'd go back to the guideline people and say, this was confusing, can you make this better? And then I think the next round of guidelines, people now started realizing, oh, we're going to have to be a little cleaner here if our guidelines are going to be used for these other sorts of methods and metrics. Yeah. So I think it's improved some. It's interesting but it's hard, it's hard to build guidelines. It's a hard project. And I you know, I had many colleagues that were on national guideline committees and they worked very hard at it. And it was a very, you know, sort of revered but also difficult and high energy.

Dr. Craig Joseph: Yeah, as a practicing pediatrician, I absolutely, sometimes, especially early on in my career, I saw a child who was there for their four-year checkup, and I would give them their four-year vaccines, you know, that had to be given between ages four and five, and the first time I found out that the birthday was the following day, you know, and the mom's like, well, what do we need to do? And yeah, that was hard. I mean, I know that there's no difference between Monday and Tuesday in terms of the child's protection against disease. But I also know that there's a school district, that's going to say you're not four, you’re three and 364/365ths, and that's not going to work. So yeah, it is difficult to kind of translate those things in.

Farhan Ahmad: Right. And one thing to kind of expand on that, and Jon will do a lot more details on this part of it is that a lot of times and this has multiple layers to it, there are guidelines that say if you have done XYZ, then you do ABC. Right. So if you've gotten a test and it's within a certain range, then, the outcome is different now as we're looking at this, like, did the EHR send us that data? It did not send that data. Was it negative or? So there's a lot of complexity like that that gets added, that we have to then as the product developers make a decision on which direction to go to, that guidelines may not be very clear on.

Dr. Craig Joseph: So speaking of the electronic health record vendors, were you dealing with more with more than one and, any pattern that you saw on, you know, how they were dealing with kind of outside parties such as yourself in terms of giving you the ability to kind of hook in?

Farhan Ahmad: Yeah. So for us, one thing, that we worked through, early on, none of the EHR vendors had the app stores that they do today. and even as we progressed, they started coming out, on the Epic EHR, we were one of the first, less than ten that they launched with. So that meant that we got to, prior to that, it was much harder to get the data, working with the customers, selling to the customers was harder because you were talking about custom integrations and additional work that their programmers would have had to do. Then once that coming up, that made things easier. But because we were still just getting started, there was a lot of unknowns that we had to work through. And similar for the EHR vendors. They hadn't really made up their minds on exactly what they would or would not allow from a technical perspective, from a business perspective. So there was a lot of, just trying things out to, to see what would work. What, wouldn't work. And then, some of the EHRs were very excited, early on. And then as they started to get going, there was a lot of stuff that they hadn't really thought through yet. So working through those with them. And then lastly, like, there's always the competition for the amount of data that we need. So we ran into a few situations where we could get, let's say, 80% of the data, but not the other 20%. And that going after that 20% was either you had to wait a bit for the EHRs to catch up on those or, you just weren't going to get it.

Dr. Craig Joseph: So complicated, huh?

Farhan Ahmad: Complicated for sure.

Dr. Craig Joseph: Yeah. That's why it's difficult to, start and maintain a company. And so ultimately, you two, you know, among others, kind of grew the company and sold it to, a big, company that does kind of healthcare content. And, you were with them for a while, but now you're moving, moved on to different things. Let's talk about those, really quickly as we start to kind of run out of time. So neither of you is really spending a significant amount of time in healthcare technology. Right? I think that's safe to say. Jon, what are you up to?

Dr. Jon Keevil: So I, came out of this experience, and the main person who benefited from that was my daughter, who had just bought a 28ft school bus and had it parked in our lawn and was looking to build that out into an RV. And so I worked closely with her to, to build out the plumbing and the electrical system and then got interested in that whole world of, electronics and solar panels and such for RVs. And so, after a year, she sold the bus and, and bought a trailer and kind of did the same thing and is now living full time in that trailer. And I built out the electrical system for that as well. And we realized that there was really nobody in the Madison area doing this sort of work much. And so I started a company called Madison Mobile Solar, where I fit RVs, I add solar panels and lithium batteries and other inverter systems so that you can live off-grid electrically and, you know, take your power from the sun and that sort of thing. So that's been great fun and sort of interesting side project. And I’m supporting another business with a friend of mine and such. So I've got a few projects going on.

Dr. Craig Joseph: Yeah that's not, that was not on my bingo card for you, Farhan, anything closer to healthcare IT for you now?

Farhan Ahmad: No, I'm actually maybe even further away, I guess. So, my brother has had, this preschool here locally, and he was always asking me to come and join him, and, and start to expand that. So, one of the nice things about it is we get to deal with more, a lot more people. And that's something that I always wanted to, keep developing on. So, since last July, I've joined up with him running, preschools and expanding those. And then on the side, I also do, aviation. So I got my private pilot license and just doing some hobby projects and simpler, projects around that, as well. So I'm pretty far away from healthcare for a little bit.

Dr. Craig Joseph: I love it, you both are, like, the standard serial entrepreneur, you know? find a problem or find something that, you think no one else is working on. And start working on it, and maybe it'll go somewhere and maybe it won't. And I think it's fine for you, either way. Well, when we end, we always like to ask our guests. The same question, which is, is there something, that's in your life that's so well-designed that it brings you joy and happiness? And sometimes this is a technology thing, and sometimes it's not. So, Farhan, why don't we start with you? What is that? Is there something or a couple things that you think are just so well-designed? They make you happy?

Farhan Ahmad: Yeah, I think, I'll do software because that's what I, spend a lot of time on. and it's a very niche thing. And it's a one and done type of thing where, like, when you're moving, I used to be a pretty big Android guy. And recently, like, in the past few years, I moved over to iOS. And one of the things that I’ve really enjoyed is how easy it is to move from one phone to another phone, where you don't even think about it. Whereas when we were, I was on Android, it was like, you got to reinstall everything. I know it has gotten better, since then, but the Android has, but iOS ever since. Like, I made the switch, it has given me a lot of pleasure whenever you make that switch and not have to worry about what you're going to copy and all that.

Dr. Craig Joseph: That, you know. Welcome to Apple fanboy-dom. Maybe you're not there yet. Maybe you're not there yet. But, one day you and I should take a trip to Cupertino and I'll walk you around the outside of the walls of, Apple's headquarters. We can try and get in. Maybe we'll get arrested. Maybe we won't, but we'll get a good story out of it. Jon, what about you?

Dr. Jon Keevil: Yeah, I think I have to go with a, a very simple mechanical device, which is in a vehicle that's water cooled. So I my first love of cars and working on cars was a 1960 VW bug that my older brother gave me for a dollar, and that was my high school car. And I learned how to do all sorts of work. That was that was air cooled. So I didn't know about these things yet. But in a water-cooled car, there is a thermostat. And the thermostat is this beautifully little simple device. It's a mechanical operating device. It has essentially one moving part. And that moving part is a bimetallic spring, where it's two different metals that expand at different rates when temperature hits them and when it's cold, it's closed. And so the water doesn't go out to the radiator. And then as the engine heats up, it heats this spring, the spring bends, it opens up a little valve and water starts going out to the radiator to cool it off. And if it cools it off enough, it cools off all the water. And then the little spring moves back and closes and it balances the temperature with this incredibly simple little use of just two metals that expand at different rates. There's no really you know, it's not electronics, there's no, you know, other sort of pieces that way. It's just this beautiful, simple little device that does its job.


Dr. Craig Joseph: That's amazing. And now I'm, upset I didn't pay more attention to physics in college. because I feel like that's, You really do need to understand, you know, the chemical properties of metals to, kind of really, truly appreciate that. 

Dr. Jon Keevil: Yeah, it's pretty simple, you know, that they, they just have two layers of metal attached to each other. And so they expand at different rates. And so when you when the temperature changes, the metal bends, because the one that's expanding faster is on the outside and it just bends it, it’s fascinating. There's a, there's a space analogous piece to this where they had to take materials for the International Space Station and they're made with mixtures of metals that bend at compensatory rates. So the lengths don't change at all. And so they have to be very careful because the temperature goes up and down hugely for those.

Farhan Ahmad: This is what I like about Jon.

Dr. Craig Joseph: Oh that's great I think well, both of those things, you know, that just moving from one phone to, to another phone seems like a really simple thing, that’s straightforward and, to your point, I've been in the Apple community for a long time, so I, I don't know how the, the other people, I guess that's the majority of the world, does it, but when I do it, it takes about 30 minutes and just happens. And I turn on the phone and it looks just like my last phone, only it's skinnier and, you know, maybe a little bit brighter.

Farhan Ahmad: And I appreciate it a lot more because I can see like the complexity behind it. And I'm like, okay, yeah, you guys did an awesome job.

Dr. Craig Joseph: and with that spring, Jon, like, my question is, who thinks this up? Like, that's it seems obvious after you've said it and how it works, but boy, that's an amazing application of that, of that kind of basic technology.

Dr. Jon Keevil: Yeah, it's a cool thing. I don't know when it started, but that's where I first encountered it.

Dr. Craig Joseph: Well, I thank you for bringing it to our attention, we've run out of time. This was a great conversation. I thank both of you. I look forward to what both of you are going to do next, which is totally unpredictable. And that's why I'm excited to to hear about it and, to to to watch what's going on so as to, potentially invest in some of these companies that you haven't even thought of yet. We'll stick around, thanks again to, to both of you. We really appreciate it. And, we'll see what's coming next.

Farhan Ahmad: Yeah. Thanks for having us.

Dr. Jon Keevil: Sounds great. Thanks.

Topics: Health IT, featured, podcast

Module heading text

Get the highest quality chemistry and microbiology testing services aligned closely with current good manufacturing practices (CGMP) for all types of products across all phases of development.

Subscribe to receive blog updates