In a COVID-19 world, Epistemix's technology could not be more relevant to the current global public health challenge. We will welcome John Cordier, CEO of Epistimex, to tell you how his company builds models that directly support the decision-making needs of its clients, including the ability to accurately represent the population of the city, county, or state of interest; the selection of parameters that best capture the local transmission dynamics; and the prioritization of interventions to be tested. This technology can make a huge impact on the current pandemic and any future outbreaks. Get more details how the technology works and its countless other applications.
Transcription:
We've been trying to make sure that people know what's happening right here in our backyard. So we've muted your microphones, and we've allowed for a chat and just ask, you know, you can ask your questions in there, Jonathan will keep his eyes on it. But also, this is not a time for you to shine the light on your own. Where's this is the time to focus on our guests, our guests and the work that they do alone. So I'm gonna bring forth john cardi a, he is the CEO and founder I believe in you might want to make sure that that's correct, john, and he is calling he is today. He's in Pittsburgh, not far actually, from where I am, I always get a treat of the fact that we're doing virtual and we're less than a mile away. So it's great that you can be here. And john, welcome to the program today. Yeah, make sure that everyone knows that you're the CEO and co founder of a pistol mix. And he is going to tell us a little bit about that. But before we get into that, we're going to talk a little bit about john and the venture. So first, we're gonna say, john, let's talk about your background, who is john, what's your journey to date, you know, academic, anything personal, you want to share, and maybe the passion for founding this company?
Yeah, I'll tie this all around the connection to Pittsburgh. So I actually came to Pittsburgh. for undergrad, I did a degree in biology degree in sociology, and was captain of the soccer team at Pitt. That's what brought me to Pittsburgh in the first place. And then I left for a bid. tried out a couple of different avenues of where I thought I wanted to land. The two areas were in cognitive neuroscience to do education policy, or to do some broader public health work. And so I came back to the Graduate School of Public Health at Pitt, where I met the two other co founders of EPA statics, Don Burke, who is the Dean of the Graduate School of Public Health. And then john Griffin stet, who was our Chief Technology Officer, and at the time was leading what's called the public health dynamics lab. And the two of them, they, they've been friends and colleagues for about 20 years, they were developing software to improve the way that public health can be practiced. And when I was kind of exploring what I wanted to really do, as an adult, and where I wanted to put my time, the the public health side, and making an impact on, you know, the larger population always tied back to health, but also social determinants of health, which could include education and economic, health and any number of other things from a policy side. And that's really the impact that Don and john, we're hoping to have through creating this, the software that we spun out into epidemics. So that's a bit about the kind of paths where we are and how john Don and I got synced up together.
So that's really cool. How long has this been going on?
So the the work that went into the software started in probably early 2000s. And we then in 2010, got really serious about building up the Fred platform. Fred stands for the framework for reconstructing epidemiological dynamics. But the cool thing about Fred is that it's actually named after Fred Rogers, another pittsburgher. So with Mr. Rogers Neighborhood, being all about improving the health of communities and improving like everyone can live in the neighborhood, we really look at Fred our platform as a way to to do that from a health perspective. So
perfect timing, john, because last Friday, I don't know if you know that we actually had Greg bear. And Ryan, I'm not I don't want to miss pronounce his last name, whose book is getting ready to drop tomorrow. And it's all about Fred Rogers and his teachings. And it's ready to drop tomorrow. So couldn't be more timely in terms of the freshness of the conversation there. So you have to get one of those books, we'll send you a link, because we're pretty proud of all this work in terms of the legacy of Fred
100%. I will definitely get that. And it'll last Fred Rogers thing I saw somebody put a collage of all of his sweater colors from 1969 to 2001. together and it was a pretty cool collage. So I'll shoot that over to you guys as well. That's,
that's great. That's awesome. So you are but you're not a lifelong pittsburgher. Right. You came here from somewhere else?
Correct. So I finished high school in Erie, Pennsylvania. I had between birth and finishing high school. 13 different stops. Both my parents were the military. So We've moved around all over. So born in Florida, then Hawaii, St. Louis, Arizona, Germany, a bunch of other places.
in Pittsburgh now for how long?
Really since 2010, with a two year gap. So I've lived in Pittsburgh longer than anywhere else. That's great.
Well, we're glad to have you because everyone, you know, put your seatbelts on, because we really need to hear about what they're working on. Because your company's really tackling, you know, some not, you know, not even just some powerful issues, but very timely issues. And these issues are confronting many of us, right, I mean, hybrid world trade shows, social determinants, I mean, we can go through almost every single thing that's manifested during the past going up during the pandemic. And I'm really thrilled to dive in. So let's just take it from the top. Tell us about the company and what you're trying what you're solving.
Yeah. So we look at large populations and the decisions that our leaders have to make whether they be our governors, mayors, CEOs, like the biggest health insurance companies are biggest financial institutions and the decisions they make really trickle down and impact whole populations, millions of people. And so what we've built with this platform is the ability to look at simulations of the impact of policy decisions. So what does that actually mean? So let's say you're Governor Wolf, right now, there's a number of different strategies to open, there's a number of different strategies on how to prioritize your vaccines, each of those is going to have a different impact on Pennsylvanians. So what we're able to do is say, Well, if we put vaccines out in this order, and now that everything's opening up, there's going to be the questions on access, and vaccine hesitancy and vaccine confidence, and what's the, like, what places of distribution are we going with, all of those things can be simulated on our platform. In addition to that, we look at things such as workplaces or schools, and the school has maybe 1000s of people coming back to it, whether they be teachers, staff, other faculty are working at the school. And we can set the schedule for a specific building. So let's say I want to open up and I want to have at least 50% of kids back in school, we can look at how can you set up your schedule so that you can get kids back in school, but also put the precautions in place to minimize infection? So if it's a workplace, we measure this on, well, how many people in your employee population are immune, whether they've been vaccinated or previously been infected? What's the overall immunity in the area that your employees are coming from? So if you're, if your office is in Allegheny County, but you have people coming from Butler, or Greene County or any werkelijk, Westmoreland, we're able to set the level of immunity in each of those places, and then say, here's the total number of people that would come together in this office building or in this school, and then say, underneath these circumstances, with the mitigation strategies in place, how many total probable infections might you might you have? And how do we minimize that? And how can we look at the staging of returns based on an observable measure, which could be something like a level of immunity or a level of vaccination? and anti you're opening plans to that rather than trying to pluck a date out of thin air and hoping that that's going to be the date that everything magically is okay, yeah.
I like to think we didn't pluck a date out of thin air. But given what you just said, I would say that we probably need some of your guidance. And it's really reassuring. I mean, what about like the companies? Like when I say thin air, yeah, we'll have like 97% vaccination and 95%. What about those companies? Is there any kind of any kind of advice or anything that you could share given that like, let's say we have 95% of our employees that have been double vaccinated, and the time has elapsed, you know, 14 days or 28 days since the vaccine, there are things that we should be thinking of, or decisions that we should be considering in terms of other variables.
I mean, really, the only other variable is, I mean, when is the the rest of the area going to be even close to 95%. And I mean, that that's an amazing number, then congratulations, that's really big. So once more and more people outside of like, your organization would be vaccinated, there's less of a risk of continuing to spread the infection even if you've been protected and are going to get symptoms. So I would say, there's a very good path of getting back into the office and getting people back together. It really went once you're going over 70% or so of people vaccinated, you're in a really good position. And one of the things that we're helping large companies with, or even cities and counties with is, well, what's the right threshold for us to begin triggering these things coming back. And one of the interesting things that we've uncovered is that the epidemic has different pockets that behave similar locally. So if you think of the state of Pennsylvania, Pittsburgh, and Philly, they are two very different types of cities, and the patterns of the epidemic and numbers of the epidemic. In Philadelphia, they resemble more of the New York, the Boston, Washington, DC, where is Pittsburgh, we talked about europen talking about Columbus, like Pittsburgh, looks a lot like Columbus, or Cincinnati. And so they're these kind of units of epidemic city that like, oh, the epidemic is behaving similarly, in these places. And we've done that analysis. And the University of Pittsburgh has done that analysis in every county across the state. And you can see, like, who clusters together. And so each of those clusters, there's a different percentage, that it's like, Alright, once we hit this percentage, we can, you know, go back to pre pandemic activities, continuing to be observant of number of cases that might arise. But with without the kind of fear or doubt of is this safe.
And so you recently won some recognition or a partnership with one of the largest, I don't want to say event, I guess, event based organizations, right. There was a press release not too long ago. Can you talk about I mean, that's really important to our constituents, right, that trade shows, events, exhibits, and not just with the tech Council, but just with all of our members in terms of them getting connected to future partnerships and customers. So tell us about that.
Yeah, so one of our one of our big partners is a company called essary. They're a big geospatial mapping company, one of the groups that actually the group that puts on their massive event, which is 10s of 1000s of people in San Diego every year, they said, Hey, do you have any mapping software, help us figure out if we can do this or not? And they said, Well, we don't, but one of our partners, epistatic does. And so we met with the CEO, and a couple other people from a company called Freeman. So they're one of the major event organizers Freeman and former Reed Clary, and they came together and formed what's called the exhibition and conference Alliance. And that's the group that's been looking at how do we bring the events industry back, that we've been providing data to them to look at, if you have events of certain sizes, with different mitigation strategies in place in locations all across the country? Because again, the epidemic is different in each place? What level and what number of people can we help bring back. So our first few events, we modeled for them were in Florida, we then wanted to, they then asked us to focus on their biggest markets. So we've been doing work for Las Vegas, and one of the big events that's a major win is the world of concrete event, which is going to be happening in June. So that's a 30,000 person event. And based on the level of vaccination coverage that the Las Vegas and Nevada should have, even though it is a national event, under the right types of precautions in place, they could safely have that event. We then started doing work for them in Anaheim, LA, New York City, Chicago, that are in conversations like the other biggest markets and they even had a couple of their partners in places like Bangkok and Abu Dhabi start reaching out to say, Hey, we want to start planning our events. Because it's such a massive part of the economy, especially in places like Vegas.
Hey, especially like organizations like ours at the tech council to so we're thinking we have an annual event and this this year, it's our 25th year of having tech 50. And it's massive. It's It's my font, a lot of celebration, amazing companies, large and small, etc. We hold it in November, usually, maybe October a few times, but November. So we're thinking, you know, we're hosting a major event in November, which we want to do right. What are some Some of the services or what are some of the things that we could get from your company that could help us there so that it would give other people who are on the call and understanding of what kinds of things that you do to make this possible?
Yeah, I'd put it into three buckets. So what's happening with the epidemic? So we can run through any number of different scenarios of what if there's another variant that pops up? Or what if the vaccine rollout doesn't go as planned? What happens if more people get vaccinated? All of those things could be really good things. That so there's the epidemic side of things, then there's the Well, what's the venue? And what are, what things can be in place at the venue to make it safer? And then the third part is who's showing up? And like, what's the age profile of that? demographic. So by was running downtown this weekend, I saw there was like a girls basketball tournament at the convention center. So like, the age profile of that audience that shows up for that event, looks very different than if it's like a medical device company event, like the age and type of people who show up with their different risks. Because COVID is very age dependent on, you know, severity of illness. So there's the what's happening with the epidemic, what plans have been put in place? And then who's showing up? And where are they showing up from? And we're able to run through those simulations to say, if we're at above 60%, immunity, and, you know, people who are showing up if you're requiring they, you know, they can check a box like, Yes, I've been vaccinated, or No, I haven't, those sorts of things, we can run through the scenarios that say, Oh, well, under situations one through five, totally safe and good to go. If there's an uptake in cases just because of seasonality, which would be like we have flu season, there is going to be flu and COVID season, depending on how that's trending, we might want to say, we're gonna have a certain precaution in place at the event itself. My inclination would be November as kind of the Right, right, timing.
Good. And we didn't even ask him to say that everyone, we didn't even that wasn't even. Thank you. Thank you for that. And we're definitely going to talk with the behind the scenes on some of this in terms of the modeling. So, you know, let's we have you know, the rest of our program, I'd like us to talk a little bit about the kinds of things that you've talked about in colleges and universities, right, what kind of work are you doing in there? What What is the future look like for them, even in the fall using that same time period? And what kind of services would you offer? And do you offer? Have you done?
Yeah, so we're looking at some of the major state, higher ed institutions. So if it's Cal State Universities, so there's, I think, 1715, or 17 different campuses across the state. And in each of those places, the epidemic is going to be at a different phase. So for in San Diego versus somewhere that's a little bit more rural, or towards the center of the state, like CSU Bakersfield or something, what we're able to do is set the level of immunity and where things are trending. And then say, if you're bringing a certain number of students back, or if you're bringing students, but half of classes are going to be hybrid, like still online, we can run through a number of those scenarios to say, here's your risk profile of the number of students or faculty. And then also, when you're bringing 1000s of people back the people in the communities that live around these universities. So there's a balance of the universities interest, the public health interests of those communities, and we're able to provide the data that they can use this more or less currency to say, Alright, here's what we're going to understand about opening this way for the fall. And what we're seeing a lot of universities considering is the you show up earlier, and then you run the semester through Thanksgiving, and then you have that extended break from Thanksgiving through the winter holidays.
So what someone's asking about the Monte Carlo simulator, you see that question? What's the base offer underneath?
Yeah, so we have an agent based modeling platform where every single person in the US is represented as a computational agent. So we've been able to build a one to one ratio of the US population we have every person aggregated to the census block group level, we have each household at the census block group level, workplaces points of interest, age demographics. And then we can overlay other physical infrastructure we can we have air quality nav, where the rivers or the roads are. So we've created an effect in the concept of like a digital twin, our digital town. Yeah. So we we've been calling, we've been using the term synthetic population for about 10 years now, we are kind of jumping on board to the, alright, when people say digital twin, we really are more digital twin like representation of a country's population. And then we can pinpoint with data from Department of Education and bureau of labor statistics we can point to here's where people go to work from the distance traveled. And so we can say, Alright, here's a workplace that has 500 people, here's where those people would come from. So we can set up the different networks between people through workplaces through even information networks, and we can overlay that on top of the population. So not MonteCarlo. But an agent based modeling platforms, what what drives all of this,
isn't it funny, it's hard to be a leader, right? All of a sudden, now everyone's calling it digital twins, and you've been doing this, this replicative work for 10 years. It's great. It's hard for people to catch up with the visionary. So it's a let's talk a little bit about vaccine equity. How about we talked about that, and, you know, the social determinants? I mean, in Pittsburgh, you know, we can make it local to Pittsburgh or anything, because, you know, we've certainly seen the disparities. So let's talk about the work that you're doing there.
So, running through a number of different scenarios on vaccine equity is something we've done with Allegheny County as our example, we've presented it to the CDC and to the World Health Organization as well. And the scenarios that we run through are realistic scenarios that a leader could make a decision on. So one thing could be prioritizing who's going to get vaccinated. This is earlier on in the pandemic, and also right now in this phase, what workplaces or worker types should people be prioritized in? What about what if we reprioritized? Age? And what if we reprioritize around specific points of distribution. And in each of those scenarios, because we have age and race, and household income built into the platform, we're able to identify with a strategy that prioritizes age or priority, a strategy that prioritizes a different type of distribution plan. How does that impact different subpopulations that are all linked to geography. And so we can point to here are the neighborhoods where we do need to do more outreach. And if that outreach isn't done, then you could expect the level of immunity to not rise as quickly there than in other areas. So we've run been able to run a number of different types of simulations around that for decision makers and leaders to say, Alright, we clearly need to prioritize vaccine confidence initiatives in these in these locations.
So there's a question in there from Justin, Jonathan, if you want to grab that.
Absolutely. So good to have you on the show. Justin, really good stuff. So given that the pharmaceutical companies are still looking for a COVID vaccine for kids of school districts reach out to you trying to figure out their next move. That seems like a logical choice to talk to you to run some scenarios about getting kids back to school safely? Yeah, so
we've done that. We've locally worked with plump school district and Riverview School District. And they've been looking at if we come back at a certain schedule or certain time, how's that going to impact the number of teachers who are going to be out and the number of kiddos that might get infected in school, and they've been able to set up a lot of really good protective measures in the classroom. Some aspects of that I wouldn't necessarily want to go to school, or you know, with like, your white borders and everything around you're kind of in a box all day, but it is it does get kids back into the classroom in a lot of places, which is definitely a good thing. So with school districts, K to 12, we're able to look at, you know, again, what's the level of immunity in the community? have their teachers been vaccinated? And what sort of protection do these vaccines give to kids and when would you might want to stage bringing younger kids back versus Middle School versus high school? And how would you want to distribute your your students coming back? Yeah. The big question related to school districts are like graduation events, or like big football games or dances, those sort of things, very similarly to how we would run an event in Pittsburgh or simulated event in Pittsburgh. So let's say there's 1000, people are gonna show up. Well, we can point to hear the 1000 or so people who would show up at a high school event, and we can add that into the daily interaction mix of what's going on in their community.
That's great. That's great. So, you know, as we wrap up, I mean, I'm sure there are people who are on this call that are very, very interested in how to leverage your services. So you'll probably get an array of interest. There is one question out there about privacy, right? Gee, what about that? I mean, I know that you're doing some digital twinning. But what about privacy,
so all of our, our whole population is aggregated to the census block level. And when we work with an organization, let's say it's one of the major event, people, we don't get names, all we need, because we already have the population represented, we just need to know how many people are coming from what zip code, and like the general age distribution, because then we can match, we can map that onto the population, and recreate that essay simulation. So we do we do not touch any individual identifiable information, then everything is aggregated to the census block group level. So that's something that also provides that additional layer of protection from that.
That's great. I mean, I'm, I'm looking at my team here with Brian Kennedy on top and Jonathan, like, we can use this, we probably are going to need to, you know, partner with you and help us do some modeling here, in terms of us testing our own hypothesis, but validating it. So we're pretty, we're pretty excited. We're pretty excited that you're actually here in Pittsburgh and doing this global work, because your expertise could certainly be leveraged to the local level. So I can't thank you enough. Are you hiring at all? Are you looking for talent, or what kind of people would come to work with you guys?
So we've, we have a few people who've come out of Pitt, or CMU, computer science, epidemiology, decision science, and even on the sociology, side of things. So we, we are all we're always looking for good people, I'd say the next phase of hiring, it's more on the sales marketing, like grow that side of the business a bit more. But we're always looking for really good software talent. And location wise, we're in a good spot within Carnegie Mellon, in town.
So is there any anything else in terms of trends that you might want to leave us with before we end our conversation today, or there's some things that you're just seeing even in the next month or so that you might want to share that might be even particular to our region or Allegheny County or something that you're working on?
So I would tie it back to the health equity side. So beyond the unjust vaccine. Beyond vaccine equity, when we're looking at things like the opioid epidemic, which is work we've done for disease control, we've done things with air quality, and where are asthma and COPD cases happening around Allegheny County? From a social determinants of health side, our platform is really good to handle those sorts of questions. And so from what is COVID taught us, it's really been a, I think, a force multiplier to shine a light on some of the social determinant health issues, not only in Allegheny County and elsewhere, but it's like the issues where COVID is a hotspot. It's linked to these other social determinant problems as well from a health perspective. So that's a whole other side of things that is getting opened up and talked about more. And we have groups like RAND Corporation and MITRE Corporation, who are beginning to do a lot of research on these topics from a policy perspective. So one thing that we're really excited about in the work that's coming up, Pennsylvania put a great tool together called the Pennsylvania Health Equity Analysis Tool. Pa heat is what they call it. And we're hoping to work more closely with that group to then say, Well, here's the statistics on health equity, as they exist today. Let's put a policy in place or let's assess the policies to see how we can look at changing these metrics longer term. And that's what we're really excited about. Good
So, john, if people want to reach out to you, what's the best way? What's the best way to get ahold of you?
I will drop my email in the chat for this group.
Great. And you know, I want it Well, as you type that in there, I want everyone to be reminded that when we have Karen Feinstein on last Thursday, I believe she's the head of the Jewish health care Foundation, she said at the end of her call, that she would be open to ideas, and for small grants to apply to the Jewish health care foundation. And they would consider and she's very, very passionate about the topics that you've mentioned here today. The organization, I don't know if you're familiar with their work, john, but and anyone else who's on the call, I just want to remind everyone that Karen did do a, you know, an extend an extension of, you know, if you have an idea, and you want to solve some of these problems, who's happy to administer some small grants? And if you didn't get that information you want it? We will give it to you. If, besides being in the chat, and it might apply to you as well, john?
Great, yeah. Then the people want to learn more about the software and other kind of far out ideas on how to apply it and things outside of healthcare. We have, like electric vehicle adoption, and how that's going to impact the energy grid type use cases and a few other things that are far beyond, you know, COVID, how COVID spreads in a population. There's been a number of other things too,
right? So this whole simulation and digital twin and redundancy, creation is really important. So john, it's been a pleasure learning just a little bit about you today, you will not be a stranger to us all. And we need we need your guidance and your leadership. And thank you for making a home here in Pittsburgh and building this company. And everyone, we are ending the show, but that doesn't mean we're not going to tell you what's up tomorrow. What's up tomorrow.
So Audrey, I'm getting getting my best beachwear out tomorrow.
Oh, right, right. Hey, that's right, Jonathan. Francis Suarez.
So you know what those guys are doing a very aggressive campaign to attract tech companies. We want to learn a little bit more about what they're doing and learn more about their tech ecosystem. So it should be a lot of fun. So
in Miami, maybe. We get Yeah, we'll have a Miami theme going on tomorrow. That's a great idea. So anyway, thank you, john. Thank you so much. Thanks, everyone, for joining us today. And we'll be here tomorrow with Mayor Suarez. Stay safe.
Thanks, Audrey.
Transcribed by https://otter.ai