Data analytics and machine learning have the potential to improve healthcare outcomes.
This episode’s guest is an absolute rockstar in the healthcare industry, we welcome Dr. Pierantonio Russo! With an impressive background in the practice and administrative side of healthcare, Dr. Russo shares his expertise in leveraging data sets in businesses, using AI and machine learning, and what the future holds in data sharing. Machine Learning has been useful in identifying and correlating different patient groups, which has proven fruitful and effective for some health systems. He also dives into how EVERSANA works and its departments, breaking down how they work in pharma without distributing pills.
Tune in to this wonderful episode with one of the greatest minds in healthcare!
FULL EPISODE
BTO_Pierantonio Russo: Audio automatically transcribed by Sonix
BTO_Pierantonio Russo: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.
Emmanuel Fombu:
Welcome to Bite the Orange. Through our conversations, we create a roadmap for the future of health with the most impactful leaders in the space. This is your host, Dr. Manny Fombu. Let's make the future of healthcare a reality together.
Emmanuel Fombu:
Good morning. Good afternoon. Good evening, ladies and gentlemen. Welcome to another episode of Bite the Orange. And today we have a very special guest. Actually, he's a rock star in a field and domain in which I've also trained in. So it's always great to speak to fellow colleagues and mentors, oeople have a lot of experiences in this particular field. If you know him, then you know he's a legend. If you don't know him yet, you're about to know him. A very unique kind of background, great, diverse kind of experiences, perfect kind of person we want to have on this podcast called Bite the Orange. This is an honor and pleasure to have Dr. Pierantonio Russo on the show today. Welcome, Doctor Russo. Pierantonio.
Pierantonio Russo:
Thank you. Manny. Thank you.
Emmanuel Fombu:
Thank you. So for those that know you and those that don't know you, and I'd like to make this about what we do as a job right now, but by the individual and then we go into what we can really do. So tell us about yourself. I know the baseline piece and that being Sicilian originally, ... and then Bologna and the whole connection on your side of it. So please tell us, what are you, what's your background?
Pierantonio Russo:
Yes. Now, a pleasure. So, yes, I was born in Italy. My family is from Sicily and my family, my brother and sister grew up in Bologna, which is north of Italy, where I went to medical school and started my training in surgery, I continued my training in cardiac surgery at the University of Padua in Bologna in, in Italy. So from Bologna to Padua and then in London. And then I finished my exposure to congenital heart disease at the Mayo Clinic in Rochester, Minnesota. At the time, towards the end of the 80s, there were only two places in the world where one could be exposed to complex surgery for patients missing left or right side of the heart, so unfortunately, children were born like that. One was the Mayo Clinic and the other one was Boston Children. I choose the Mayo Clinic because of the volume of work and the results of that. After Mayo Clinic, I took a job at Temple University in Philadelphia as a director of cardiac surgery, pediatric cardiac surgery and transplantation. My team at ... Children in 1992 was the first to operate successfully on a newborn with hypoplastic left heart syndrome in Philadelphia with transplantation. So we transplanted that child. In 1994, we were among the first to transplant a very young newborn, three and a half days old. Child did well, again, born with congenital heart disease. And many of these patients are still alive. I also ran research laboratories focusing on the, on model of the fetal surgery, as well as the protection of myocardium and the brain, and you need a model of cardiac surgery. After my cardiac surgical career, I ..., so in 2008, I started to slow down my practice. I continued to operate for advocacy in which, with a charity group called The Gift of Life International. So I operated for free essentially to teach in developing countries, including India for several times, in Chennai, and had the opportunity in India also to teach the, an MBA course in health management at the Indian School of Business on behalf of ... School of Business. But my day job started in 2008, was, with the industry, I was an executive at two National Health plans, one Independence Blue Cross in Philadelphia and the other one Pilgrim in Boston. During those years of working with the health plan, I focused my attention in learning how to use predictive analytics, machine learning for supporting various applications needed to promote value-based care, case management, targeted case management and precision medicine. So who are the patients with heart failure who really need the hospital at home? They need a visit by a nurse, they need remote patient monitoring, not all the 20,000 that we had. So we use machine learning to identify the rising risks. Those patients that today they seem to do okay, but in reality, in three, four months, six months from now, may end up with multiple hospitalization and for various reasons, including most importantly, social determinants of health and other barriers to care that we know are unfortunately part of our multiple communities in this country. So at the, in Boston also was the next general population medicine ... At Harvard Medical School. And I had the opportunity to support my colleagues in medical home to build up the hospital home program, essentially think of the teen conditions that instead of being treated in the hospital, are treated in the home and that your hospital role is now in your home. So you see your doctor virtually, but you have nurses all around you and specialized individuals who can bring in radiology, lab tech and, and ultrasound, infusions, anything that you need. This is a very important new opportunity for improving the care of particular patients with chronic conditions who require multiple rehospitalization. And then in 2019, I joined a company, an analytic company called HGH, now is part of EVERSANA as a chief medical officer. So working on using data in analytics and most importantly, machine learning for various applications. Now, EVESANA has more of a path towards supporting pharma and device companies, but essentially my role is a corporate chief medical officer of EVERSANA is to support all the departments that are part of EVERSANA. Think essentially EVERSANA is like a pharma without a pill. Every single department ... Pharma would have, a diverse company would have, from medical affairs to analytics is there. And my job is really to make sure that we translate clinical knowledge into analysis that our clients and our physicians in the community can utilize and make it a positive ... The other thing I also work a lot is on rare diseases with the NIH, where we are studying the way to reduce the time to diagnosis of rare diseases by utilizing machine learning. And now we are working on computerized phenotypes that can support physicians in the community who are not familiar with rare diseases to alert them that this particular patient needs elevation and needs a genetic test or needs an enzymatic test, or even just refer to a specialist.
Emmanuel Fombu:
Doctor Pierantonio, it is quite fascinating. I mean, like for my background kind of perspective and to where we are now, on the use of datasets, right, ... patients. You mentioned the NIH program, for example, especially around rare disease, actually have a study that I ran recently in Germany at Mainz and Ettlingen in Germany using, it's a clinical decision support program that we built to identify patients with Fabry disease within these hospital systems ... the American consortium piece of it. And people kind of miss how highly important and critical it is to identify these patients early to actually, you know, put them on therapy much earlier and have much better outcomes on upfront. But with that being said, and we talked about that piece of it. So what's your why? What got you very interested, first of all, back to your origin piece. What got you interested in cardiothoracic surgery? As a young kid growing up, what was the fascination? What pulled you towards that particular aspect of medicine?
Pierantonio Russo:
Yeah, I still remember when Christiaan Barnard did the first heart transplant and of course I was a kid, and was fascinating to see how medicine now was using technology in order to improve outcomes. And so to me, cardiothoracic surgery is always been, even more than neurosurgery, being the path of medicine that uses a lot of the physics and technology. So physiology, if you don't understand physics, you cannot understand physiology, right? And if you don't understand physics, you cannot run a cardiopulmonary bypass machine or working with your perfusionist to, to, to work in a machine, ECMO for example. ECMO, we didn't have these small artificial hearts, of course, ... the ventricle system ... in children. And so the only mechanical support that we had for a newborn was ECMO. And now, of course, we have the ... Heart and other things that can be applied in small children. But, you know, in 1990, 91, 92, 93, ECMO was the only thing. And now, of course, ECMO was used as a bridge, to bridge the bridge to transplant. So the integration of that basic technology and medicine was what fascinated me and what made me get involved. And then, of course, I was always curious about new things, right? So fetal surgery, the bloodless surgery, how we reduce the signs of an oxygenator in the heart machine in order to avoid, for example, blood transfusion in the children of Jehovah Witness, which would, Philadelphia, we had plenty. And I'm proud to say that the Jehovah's Witness Church took us on as the, their advocates in Philadelphia without court orders, all the other things that happen all the time because we knew how to take care of these patients appropriately and respecting their preferences.
Emmanuel Fombu:
Which is quite interesting, and that goes comparison. I did some work in nitric oxide as well, I used inhaled nitric oxide as a bridge too, right? Well, right there. And the impact of that, especially around Covid, ... and I just working in ICUs, which is interesting. So it's great to have that background piece. But we fast forward to something that, I saw an article that you read that I found quite interesting .... I was about to use some machine learning to accurately size market potential and optimize sales and marketing resources, right? So a lot of the audience that we're listening to, a lot of entrepreneurs, they have all these stand up companies that could be in the idea of what is clinical decision support or is identifying the right hospitals to sell their solutions to. And you have great data analytics capabilities within your company, within EVERSANA. So it's quite fascinating. So with that being said, I want to look at value-based care. For example, I've had this idea in general that if we want to help companies, let's say, have an app for diabetes or an ... and heart failure management or the app for disease management, identifying patients. If I go to health system usually is how do you, what kind of data sets do you need, right, to actually make it, to identify where to go, who to talk to, to get the most value out of it, right? So once you develop a product, is a key element difficult to do a lot of startup companies face about identifying with the right customer is and what is that value proposition and what like API to measure. So, so tell me some of the ... insights and I saw some of the great things that you heard about and how you leverage EHR data, some of the commercial data sets, rural evidence, to fine-tune in. So what is that roadmap, as you know, the chief medical officer, someone with a great background, as you have, if have a solution that want to sell to a health system, you can use any example you want, what are some of the data sets that are available and what is the best way you think they can be leveraged to make sure that we actually launch products and adopt and scale some other solutions?
Pierantonio Russo:
So the next in question, so currently I would say that the electronic medical records, the clinical registries, the data from remote patient monitoring and wearables and of course claims are the traditional data sets that even the FDA considers as real-world evidence. I'm going to discuss a little later, though, my view that we need to move to the next new thing, which is data sharing. So the Mayo Clinic platform that allows to share data and essentially take advantage of a simple concept. Healthcare is an ecosystem, and that ecosystem, there are plenty of players. Each of them had their own modules that can contribute to the ecosystem. But obviously when you have high modularity, you need a coordinator, you need somebody who can facilitate the data sharing. And I think that there are roles for people at EVERSANA, so companies like EVERSANA, companies that you may know that they act as not the facilitator for that data sharing that indeed provides the opportunities for patient support group, patients, pharma, the PBMs and physicians and health systems to actually find a way to, in a safe, in a privacy sensitive way to share contemporaneous data because we all know what the gaps are in the electronic medical records, right? I mean, you have plenty of residents who run around in the copy and paste the same notes from the day before, who have incomplete data. Now, of course, I welcome the opportunity that Apex is giving now to some of the facilities where they have electronic medical records to utilize this generative AI, essentially using OpenAI to listen to conversations with physicians, between physicians and patients, and nurses and patients, social worker and patients, and then translate that conversation in a comprehensive notes, right? Because since the, 2014, actually in 13 at Independence Blue Cross in Philadelphia, we use social determinants of health to data. There's many domains that we could find to improve the accuracy, the predictive power of our predictive models. If you have a patients with heart failure who is socially isolated and that lives far away from a medical centers and need a big bus to go to the hospital or to the doctor, that patient is going to be much easier for that patient to call 911, even if the patient is on the early stage of heart failure, while the patient who has plenty of family support and hence a physician who can go to the home, who is not going to go to the hospital even if he's needed to palliative care. So .... is extremely important. And where do you find the evidence in social determinants of health in electronic medical records? Only when the social worker is diligent enough to put it down, right? But now, if generative AI allows you to listen to that conversation, then it is translated into a comprehensive data, the comprehensive note. And I think that the data sharing platforms will offer an opportunity for true collaboration. Meanwhile, I would say that machine learning has been used to build up the appropriate cohorts of patients of interest and of course validated cohorts. And then of course the various algorithms, machine learning algorithms, identify similar patients in databases, mostly in claims databases, but also in electronic medical records with NLP that allows you to identify those patients who have been not being diagnosed correctly with being undiagnosed or not being appropriately treated. So even for quality purposes, you know, you can compare the care of a patient and the patient journey with recommended compendia and professional society recommendations. Now, it is a tremendous opportunity for quantifying the poor value care that we unfortunately still part of our healthcare system. We applied the recent, we are going to publish a paper on the use of machine learning for identifying patients with hepatitis C much earlier and with much more focus than the usual screening that the CDC is offering. We're also publishing a paper on the use of machine learning to identify the most probable treatment that prevents the progression of MS, from relapsing MS to secondary progressive. I just presented a paper, The Digestive Societies in Chicago on the use of machine learning to identify patients with gastroparesis, even if they don't have a labeled gastroparesis. So there are plenty of applications and we need to continue to use machine learning on the appropriate set of data with appropriate clinical domain expertise.
Emmanuel Fombu:
Which, Piereantonio, as you mentioned, is quite fascinating, all the different things that you can do. And so EVERSANA, as an entity, is quite massive, right? And so I think having you in there, most people listening and with ... You've had in there and experience, although you mentioned earlier that it's like a pharmaceutical company with that appeal, but I'm sure you're also working device companies, I've seen some of the great rural evidence pieces that you've done, but because you don't have a pill, you actually makes it more flexible and things like you do, right? Because you have more trusted as a partner, right? In that sense, where you're not doing this because you want to push any particular drug to anyone, right? You're actually making healthcare in a much efficient way. So for those that don't know and anyone listening how to partner with you, of course, at EVERSANA, what is the overall kind of structure or the way EVERSANA is right now, you have different departments. What are some of the services that you offer? What kind of clients do you have now? And yeah, I think have a good start for someone listening to learn.
Pierantonio Russo:
Yeah. So EVERSANA is a 7000, employee-strong company, is a global company. We have offices practically in, I would say every state. And we do have a presence in Europe and also in Asia, in India and in Switzerland and in London, and we continue to expand. Essentially, it is a company that allows end to end commercialization. So think of the, all the companies that decide for various reasons to outsource the commercialization from prelaunch to launch and implementation. And we have, of course, medical affairs, we have compliance. We have a specialty pharmacy which is associated with our patient services, patient services, which of course allows to provide support, whether it is economic support, coupons, pre-authorization, but also continuous support from the standpoint of persistence and adherence. We have, again, the commercialization with activities that allow to launch a new product with experts that hire the appropriate personnel from rep to liaisons with hospitals and other organizations. And of course we have data analytics and we have consulting services and we have an agency. So at the center of this is data and analytics, because every single department is using data and analytics in order to support the primary research with actual transactional data. What it really happens in clinical practice, in the care delivery system and the machine learning team, the data science team obviously makes sense of that data in order to create the opportunity to identify trends and making sure that all the dots have been coordinated and identified as one story. So if we put all the dots together, what is really the trend for the physicians who are wanting to use a new drug but they cannot use it? Well, they cannot use it because market access or physical, patients who may benefit from that particular drug, but unfortunately, they have issues with copay or the insurance that decided, their insurers decided not to cover. So the mosaic of data put together in a way that actually the final analytics is credible and it is actionable.
Emmanuel Fombu:
Which is quite fascinating. I really like this approach because you have a complete end to end kind of solution, right? It's an end to end informal idea of thinking through a process, execution to the process to actually tracking the outcomes of entire process. But with that being said, there are many challenges in healthcare, whether it's across therapeutic areas, whether it's a cosmetic first, whether it's R&D or commercialization, product launch many different angles. So in your role right now, what are your top main focus kind of areas that you are? I know oncology is a big piece, right? This is a big piece for you guys and next in neurology. But what is, this is one of the key areas that you're focused on in the short term, right now?
Pierantonio Russo:
So practically every therapeutic area in pharma, from oncology, cardiology, neurology and immunotherapy, we also have a digital medicine decision department. So the medical, digital therapeutics and digital medicine, as you know, is a new area. Do I take advantage of data, but also of technology. And so a virtual care is here to stay, we support telemedicine and other digital therapeutics opportunities from various companies, and we have a strong focus now on the generative AI to create medical education besides the agency medical education, because of course you can create a twin of a doctor, a twin of a patient with certain characteristics that they can interact together. And of course, my next dream in a way is for our generative AI to create also metaverse solutions where the patient and their doctor can work together, an alternative reality, and in my opinion, having worked in population medicine, I believe that when the patient actually imagined themselves and then see themselves in this alternative reality with all the options that that come out of the choices that they make, then they become more aware of their condition, aware of the need for wellness, and most importantly, they become adherent to the therapy.
Emmanuel Fombu:
Again, the question is what have you been my whole life, Piereantonio? I'm glad we're actually connected to this way. And I was talking to a good buddy, Alan Gilbert, actually recommended us.
Pierantonio Russo:
Yeah.
Emmanuel Fombu:
A physician ...
Pierantonio Russo:
Yeah. So, so how did you come in? And it's interesting. I just texted him right now as we're talking, right now. And I sent him a picture of us. I was talking, it's quite interesting when you talk to people and you said this idea, so me talking to you right now and you talk about the metaverse and things that you're doing, it's exactly the things that we're passionate about, especially in the world of mental health. But Pierantonio, we imagine people having therapy in this virtual metaverse worlds, where people can be themselves and they're taking away these barriers and they could create this personality of who they really are without being worried about being judged and these other world. And I think that the concept of a digital twin, the clinician perspective and I like what you did, I don't even think that far ahead of it myself, right? Because having the virtual patient, the virtual doctor, communicate, I never thought about that, right? So something I learned from this, I think it's quite fascinating. So expand on that a little bit because I think it's well thought out. And for someone that's listening, what is this virtual world look like, right? This world? The virtual doctor, the virtual patient, what is that, yeah, why do you care about that? And what are the opportunities that that world opens?
Pierantonio Russo:
Yeah, I think there is a company in France called Accenture, and this company essentially has already experimented. It's been experimenting on creating those digital twins of a patient with all the characteristics of their .... That one is targeting. And they introduce in that simulation all the clinical signs and symptoms of the disease, but also the metabolic characteristics of the patient and even the enzymatic environment like the PY260 enzymes that metabolize drugs. Some people have a lot of it, some people have very little of it. So they then simulate the response of that particular virtual twin to a certain drug with certain characteristics. That to me is huge. And again, they just started with this work. I think we are watching it. That is for the drug development. But from the standpoint of the patient experience, I think it is creating an augmented reality or augmented reality that allows you to immerse in the experience from the time you are in Avatar. Now you are using Avatar as a patient and as a doctor. The first interaction with the doctor, the first interaction with the pharmacy, and then you go to the pharmacy and you discover that your prescription is not being filled because the insurance company has denied pre-authorization. So then phone calls, your doctor needs to call the pharmacy. So sometimes it takes up to three, four weeks for patients to get the medication and then you are starting to take the medication and then you have some collateral effect that your doctor or your nurse didn't talk to you about, so then you get alarmed that you go to the emergency room. So all those things that actually happen in the real world allow us to do two things. Number one, to learn more about the patient experience, then learn more about the patient own preferences. And finally, all of this may allow us to, number one, reduce the friction in the care delivery system, but most importantly, be able to will allow patients preferences.
Emmanuel Fombu:
Which I think is very important. That kind of personalized medicine where is geared towards the individual, right? Because every patient is unique, their response to a drug and this comes with diagnosis as well, right? So how you identify patients, which I really like, what you work right now at EVERSANA, where you have real evidence of data sets, you have all this different kind of worlds that you created where you're personalizing and making sure that healthcare is unique for that particular individual. But with that being said, so what do you see as the main challenges right now and what are some of I mean, of course there are always challenges, but what are those immediate challenges that you think our industry faces right now?
Pierantonio Russo:
Well, of course, the healthcare, in general faces the problem with inequity and the lack of access to appropriate care for many individuals. But thank God, at this point in time we are trying to convert collectively I think, we are trying to address that even in clinical trials, in designing clinical trials. Second, I think that still patients pay too much for their medication, even complex diseases associated with a reimbursement structure in terms of medications and treatment that unfortunately doesn't cover all the financial needs of our patients, and so we need to resolve that. There is, of course, the price of medications that continue to be a problem. And, you know, if you listen to the industry, it is the PBM faults. If you listen to the PBM, it's the industry's faults, but one way or the other, we need to resolve it. And of course, price transparency is the way to go. Let me know if you have paid rebates to providers and payers, let us know, so that then we can identify why the prices continue to go up, if there are obviously. Now, I think that there is one major bright light and that is the understanding of the concept of value-based care. So a care that really provides, if not cure, improvement in patients lives and that value care can only be achieved with more precision medicine, and that precision medicine can only be achieved with strong use of data, multiple data, multiple sources. And of course the way you using that data.
Emmanuel Fombu:
This is quite fascinating, Pierantonio, I'm sure we can have a three hour conversation on this. I would love to definitely invite you again to pick specific topics. I'm sure there's several angles we could talk about, why .... which is good to have this chat because I think that the multiple angles and flush out those in more details and specific verticals. So this is like the initial tease for everyone listening, ... will love, you know to definitely have you again on the show but thanks for your time today, I don't want to hold you up.
Pierantonio Russo:
Thank you.
Emmanuel Fombu:
Long and hopefully we'll get to see you again.
Pierantonio Russo:
Absolutely. My pleasure. Thank you so much for inviting me.
Emmanuel Fombu:
Thank you.
Pierantonio Russo:
I really enjoyed it. Thank you.
Emmanuel Fombu:
Thank you for listening to Bite the Orange. If you want to change healthcare with us, please contact us at info@EmmanuelFombu.com or you can visit us at EmmanuelFombu.com or BiteTheOrange.com. If you liked this episode and want more information about us, you can also visit us at EmmanuelFombu.com.
Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.
Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.
Sonix has many features that you'd love including powerful integrations and APIs, enterprise-grade admin tools, collaboration tools, automated subtitles, and easily transcribe your Zoom meetings. Try Sonix for free today.
About Pierantonio Russo:
Pierantonio Russo, MD, FCPP, FAAP, STS is a CT surgeon and healthcare executive, currently in the role of Corporate Chief Medical Officer at EVERSANA. He has practiced cardiac and heart transplant surgery for more than 20 years. More recently, He has held executive positions in data and analytics, predictive modeling using ML, health economics and reimbursement, health insurance, and population medicine.
Things You’ll Learn:
If you don’t understand physics, you cannot understand physiology.
Data sets from wearables and remote patient monitoring devices are a gold mine.
Healthcare works as an ecosystem.
AI and Machine Learning tools need to support the jobs done by humans..
Digital care is here to stay.