Note: The I AM GPH podcast is produced by NYU GPH’s Office of Communications and Promotion. It is designed to be heard. If you are able, we encourage you to listen to the audio, which includes emphasis that may not be captured in text on the page. Transcripts are generated using a combination of software and human transcribers, and may contain errors. Please check the corresponding audio before quoting in print. Subscribe now on Apple Podcasts, Spotify or wherever you get your podcasts.
EP128 Harnessing Data Science for Cardiovascular Health with The Novartis Foundation
Aman: Folks, welcome back to another episode of the I AM GPH Podcast and today we're diving into a groundbreaking project, that's set to revolutionize the way we address heart health disparities in cities, starting with New York City. The AI for Healthy Cities Health Equity Network is a collaborative effort initiated by the Novartis Foundation, in partnership with Microsoft AI for Health and the NYU School of Global Public Health. We're on a mission to transform the landscape of public health. This ambitious initiative leverages artificial intelligence and data science to provide health authorities, health system leaders, and decision makers with insights and tools to tackle the root causes of heart health inequities. Heart disease is the world's leading cause of death and a significant driver of health disparities, particularly in urban environments. But this project isn't just about data and algorithms, it's about harnessing the power of interdisciplinary collaborations to address the complex issues at their core, which is why we have two amazing guests on the episode today. We have Elizabeth Adamson, the Associate Director of Population Health at Novartis Foundation and our very own Dr. Jose Pagan, the chair and professor of the Department of Public Health Policy and Management. Dr. Pagan and Elizabeth, welcome to the IM GPH Podcast. We're so glad to have you here, to talk about this fascinating topic.
Jose: Thank you for that introduction, man.
Elizabeth: Yeah, thank you for having us.
Aman: It's a pleasure. Now let's get started. I wanna know, what is your personal motivation, behind this work? Why is it so important for you two to do something like this?
Elizabeth: Personally, I feel incredibly privileged to be working on a project which has the potential to do such good. I feel really lucky that I love what I do and have the pleasure of working with such dynamic and diverse individuals, such as Jose and a number of our different partners, in not only in New York City, but also Singapore, Basel, Lisbon, and Helsinki, we have other AI for healthy cities, across the globe. And one of the interesting things is, personally I have few items of experience in data, but I'm not a data scientist and so my personal motivation is around the impact and the potential for impact. And I grew up in Washington DC, I went to University of Edinburgh in Scotland, I lived in and worked in about seven different countries before the age of 25. So I have a lot of experience in trying out different roles and responsibilities, from archeology, human rights, teaching younger kids and came into this role, because I experienced a lot of personal growth from those activities, so when I worked in human rights, I had, sort of, the sexier title of working in human rights and making a real impact and it was a great dinner conversation. But when I thought about my reach, my global reach, it wasn't on the systemic level. It wasn't working within an already established system like a health system and really changing the way it's done or works. And so, when I joined Novartis eight years ago, I started by working internally to try and transform the way we look at development, of not only medicine, but treat disease and realized you have to work within that system to have the change that I want to see. So I personally believe that there is a brighter future, there is a more equitable world, and I wanna be a part of the people like Jose, NYU, Cornell, Microsoft, that are trying to chip away to make a more equitable world.
Aman: Wow! Systemic changes to have global impact. And we can see that passion from your upbringing as well, in multiple countries, you have been. This is going to have the kind of impact, I mean, wow. Dr. Pagan, I'm excited to hear your thoughts. What excites you about this?
Jose: You know, when I started doing research in health, the topic that I was most interested in was how does context impact people. Like, for example, I was doing it on insurance. So it's about 20 years ago, I started looking at this issue of like, if I'm uninsured, bad things may happen to me, because I don't have access to care. But then I started thinking about, and working through with mentors and so on, on this topic of like, if I live in an area, where everybody's uninsured, what impact does that have on everyone that lives in that area? And back in the day, when I started doing this work, there were many, many problems that you ran into. One is not having data, another one is your computer models will take days to run. You did it, but you did it with the limited resources that you had and even though you had those constraints, you still did the work. So, to now, when this opportunity came up to work with the Novartis Foundation and when I saw the work they were doing and our colleagues at NYU saw the work they were doing and we met with them, it opened up a world of possibilities of how to impact health, in terms of how can you put together data in a city, to inform cardiovascular disease, hypertension, all these indicators that matter. And it's, how do you do it, working with technology companies, the private sector, how do you do it working with the foundation? How do you do it in an environment where you may learn or you will learn something from other cities around the world, that may be doing similar work? All those pieces, when I did this 20 years ago, were not there and it becomes frustrating over time, so when you have a chance to do that, it opens up many doors. Getting students involved in the project is a key part of it. And, then more recently, the experience we have with covid and what we saw in terms of health disparities and so on, added to it and made it even more relevant. So anyway, that was a lot, I know, but it has been, what, a couple of years now, maybe a little bit longer? And it has been a very productive partnership.
Aman: It wasn't long, Dr. Pagan, in fact, it wasn't enough. There's so much more to hear and I'm curious to know, how did this collaboration come into play? You mentioned the private sector, the academic sector. When I see a project like this, I see a lot of cross collaboration. So how did the Novartis Foundation get connected to NYU? How does this even take part?
Elizabeth: First of all, Jose, I have to say it has been less than one year.
Jose: Less than one year. Yeah. I guess I was counting the first time I started hearing about it and yeah, but you're right, we've done a lot in a very limited time.
Elizabeth: We've done a whole lot and we've learned so much, throughout the process and I think that this is also why someone who is a non-data expert, working on data and technology and analytics, while we have technical expertise within the foundation, there's huge opportunity for traditionally non-tech to get involved in these sort of activities, because there is so much more needed, different skill sets that could be applicable, because there are so many interwoven synergies between health and technology and we need to embed it more and leverage the expertise from other disciplines, to learn how we can move forward. So how did this initiative start, Jose? You wanna tell?
Aman: Yeah, tell us about the partnership. It's such a cool partnership for academia, at least for us at GPH. We like hearing about these stories, where inside such big projects that are happening, how does this even come to play?
Jose: I don't know, we have faculty that do work in different places. The foundation was interested in New York City, faculty here were interested in working with the foundation, so basically a group of us work together with the foundation to make it happen and the way it works is you do a lot of back and forth with any foundation, not just Novartis foundation, about what are the priorities? What they want to do? What is our capabilities to do the work and what other partners are needed? And before you know it, after several months, you end up with a project that is mutually interesting. And in this case, the project that we have been working on is connected to how can you learn about how you can integrate data, to impact cardiovascular health and disease? And also, how can you identify the data, integrate it, get partners at the table, in a place like New York City? And that's what we have been basically working on. And then the project gets connected also to other activities that the foundation does.
Elizabeth: So the foundation is focused on improving population health for lower and middle income countries. And we focus on cardiovascular disease, because over time you can measure the impact of anything that you change on cardiovascular disease, because there's such clear clinical markers for it. And so what's really interesting, to me, about how the partnership began, and I think the motivation it has for future partnerships, is because this started from an idea and a lot of times you don't hear that evolution, you hear the struggle or you see the end product, but I think what students need to understand is that this is how partnerships and how these bigger initiatives are formed and if you never try or you're scared about what the other partner, what their incentives might be, that's gonna be a huge barrier to innovation or change. And you really need to, at least for us, I think we took a leap and Jose said in the prep call here, he spoke about how any relationship, it's like any other relationship, you give a little bit, you test it out, you build trust. And so along with the learnings that we've had for the initial program, what we hope to build is beyond that initial success. Because as we go forward, we're gonna have a newfound trust, to take bigger leaps on bolder ideas, understand where the other organization is coming from and better meet the expectations from all ends. So I think, while it's tough, if it was easy, I think everyone would be doing it, because it will amount to an enormous impact.
Jose: It's interesting, you can pick any problem, you wanna solve that may seem very difficult and sort of like, as long as you bring folks with very different perspectives to the table and that includes folks in the private sector, public sector, students, like any voice, and then you have to have a structure, like you have to have a… somebody told me this about a project once and said, “all you have to do is meet once very week at eight in the morning on Wednesdays, and your project will go just fine.” And in a way we did some of that, that we had a structure on how we met and we faced many challenges along the way, but then through those conversations, you start trying to figure out exactly how you're gonna move forward. Like, for example, how do you get data from a hospital system or how do you get data from certain agencies? Or how do you collect the data and put it in one place? Who owns that? All the legal assets.
Elizabeth: And what data do you even need?
Jose: Yeah.
Elizabeth: What data do you even need to get to where we want to go.
Aman: Let's talk about data, there's so much there. I'm very curious to go specific into this project, there seems to be this structure. You've given us a blueprint of what it's like to collaborate and build something upon that structure. I'm surprised that it feels like you've done all of this in three years of work and just discussed that and yeah, it's been under a year. Think about it. So much of the consistency that's taking place, I love how, what are the real world impacts of this? You know, the interventions on a population health level, the partnership is definitely gonna have a big impact for New York City. What is the problem you all are focusing on right now and what is the intended outcome of this? How does this outcome take place in the process?
Jose: I think being able to have, I'll give a brief description and then, I think being able to have data at a certain level, census track level and so on in the city, about social determinants of health, different factors that are connected to health and the health outcome itself and being able to have the data over time, be able to visualize how that data looks like, and then use it for decision making, is what we're trying to get at. And a lot of folks can use it, it could be a health system, that's sort of like, the easier way of looking at it, because you know they're dealing with healthcare and health. But it could be a community based organization, it could be the city, it could be anyone and the challenge is, how do you take all that information connected and then provide it in ways that are useful for folks to use.
Aman: You know, I'd love to pose this question to Elizabeth. Since you're coming from the non-NYU system right now, there's a lot of people that watch this podcast, most of them are young public health professionals. They're either undergraduate students, grad students, PhD and everyone trying to find their future opportunity in this world of public health. What is their niche in public health? Where can they find their public health knowledge, their gifts that they are to give? What are the future career opportunities in the world of health and tech that are coming up? You're all doing some amazing projects right here and I'd love to hear about the other things that are happening, that students could be aware for, when it comes to their own future and the future of public health and tech.
Elizabeth: So I would love to answer that question and I will in one moment, because I just wanna respond a little bit, to what Jose said about the potential of something like this and what the real world impact will have. And this is also in the direction that the world is moving. So there's a lot of hype right now, around generative AI, ChatGPT, a lot around some of the ethical challenges, with image recognition, bias and algorithms for artificial intelligence. And what we're finding is actually, or at least my personal point of view, is that, while I think the real impact is gonna be in predictive AI, once we can get that right. And while generative AI is fantastic, predictive AI will be a tool that humans can use to make decisions better. And it's not to say that this is an easy task, and it's not to say that the tool will make the decision. It will be humans that sit together, analyze the options and make a decision on the best approach, using prediction models. So what this project has the potential to do is to put these decision making tools in front of the community, the public, government health decision makers and actually enable them to make better decisions. So instead of being as reactive in a healthcare system, to health threats from let's say climate particularly, which is literally boiling the news right now, can we make more proactive decision makings to improve our health over time? And can we simulate that impact through, so when we say, okay, let me try putting this food truck in this neighborhood versus this neighborhood, because I have limited resources. If I put it in one location versus another, you would have X amount more impact on population for that neighborhood, than if you did in the other neighborhood or it helps resource planning, resource allocation and ultimately, this is the direction I believe health is going in, is around prevention, because we're having higher healthcare burdens, higher healthcare costs for government, they're trying to figure out ways to reduce those growing costs and prevention is a long-term strategy that previously wasn't as possible, within short term government horizons of elections, et cetera. So the future is around understanding what the actual cost is, of preventing disease and showing those long-term benefits and quantifying it, so governments will make better decisions, that in the end will keep people healthy.
Aman: It's so reassuring to hear the possibilities. It's inspiring, in fact, to see that there's something that predictive AI can do. This is not something we're talking about, but the impacts that it could simplify certain things that may not be easy to understand and that AI can take care of that is simply fascinating to me. And let's pose that to students as well. What can students do with this? So there is so much more beyond predictive AI, as well as in tech, what are certain industries that you have identified in all your experiences and we have seen in the health and tech field right now?
Jose: Anything that allows them to work in teams and communicate and get exposed to different technologies is gonna be important. So if you can spend the next two years of your MPH or any other degree you're doing, learning these things and being exposed to people that are using, it's important. So I tend to think about it, in terms of skills that you need to accumulate while you're going to school and they tend to be then, teamwork, but also knowing anything connected to different technologies, coding, data visualization. You can pick whatever aspect of it that you're interested in and learn it. So for example, we have students that get interested in health insurance data and it's very specialized but you need to understand how the data is structured, human error into the data, how do you deal with it? And the students sort of like self-select themselves. You self-select yourself, you become all about data visualization. But some people get into all about predictive models, to look at cost, social determinants and cost, within that type of data. So the more that you can learn about these things and learn something substantive, not just do one week of this, another week of that, two months of this, many students do that by the way. It's better if you do one thing and then you can explain it to an employer. I was part of the AI project…
Elizabeth: That's a great point.
Jose: This is what we did. This is what we did, here's the product.
Elizabeth: Stick with it.
Jose: Yeah. This the product.
Elizabeth: First, job three years, I was told was one of the advice.
Jose: You worked three years?
Elizabeth: Three years and follow one project, start to finish.
Jose: Yeah.
Elizabeth: So you can talk about it, so you can speak about it, so you have expertise and then follow your curiosity.
Jose: Yeah. Have a technical report that you did, cause then that tells the employer a lot about you. It tells you that you have a sense of beginning to end on how a project works. And a company. And you know, I get this question a lot, Elizabeth. Like I get students that tell me, "I got two Summit internships, should I do one month and another month in like two different companies?" And I say, "No, no! Pick one you like and stick with it."
Aman: Not to make this about relationships, but it's like choosing your partner, right? Is that what you all are saying? It's like managing so many different things, that you can never invest fully to some degree.
Elizabeth: But I wouldn't worry so much about the choice. I mean, not in real relationships, the choice is very important! Rafa, if you're listening, very important! What's most important, is showing your values. And I don't think that you adapt your values to your organization. I think there's a real opportunity, which I didn't realize going into the job market and took me a long time to figure out, is you have a lot of flexibility in your role. Yes, you have to do the top three things they hired you for, yes, you have to work within those boundaries, but how you get there is up to you and if you can really shape a role and put your passion in it, they actually want you to do that, so you're more committed to your job. But it's up to you. And I think if you surround yourself, in terms of the choice, you mentioned choosing your partner, and I think if you surround yourself with people that inspire you, that challenge you, it can be in any field, even if it's not necessarily your core focus, and that's what I'm finding also with data and technology, I didn't know that I was going to go in this direction, but because of the motivation, because of the potential of it, I know more about prediction modeling than a lot of people that actually specialize in this, about the landscape of it, about how to actually integrate it for action. And so I think you need to also find your niche within that role. It like, how can I add value to this project? I might not be the technical expert, but how can I create value for this group and what's my role and what's other people's roles?
Jose: I've noticed that many students worry about like, "But I don't know anything on a specific topic," or something like that. They worry a lot about things like that and nobody knows everything about everything, nobody does! So the moment you realize that nobody is looking at you as somebody that knows everything, you become more relaxed and you know, just, I think that as long as –
Elizabeth: This would be another advice I have, actually, is say when you don't know. Don't pretend. But if you know what you don't know, you're in a very good position. So figure out what you don't know and be like, "Hey, I don't know, but I'll figure it out." Or, "I can find someone who does."
Aman: Yeah. Yeah. I just imagine Dr. Pagan in office hours, where the student says, "I don't know this!" And Dr. Pagan replies, "Well, yeah, nor do I."
Elizabeth: He did that to me!
Jose: I'll give you a good example. Students get invited to interview sometimes and they freak out, 'cause they're like, "But I don't know this." And I'm like, "Don't worry about it." Or they get hired, if I hire someone, I already know that person's gonna be able to do that job. So you may not feel like that, but they know you're gonna be able to do a job. Like, a lot of it is in your head, you know?
Aman: Yeah, yeah.
Jose: Barriers, you put yourself. Do it. And at least I gave you a a great example of how, there are different pathways to get to the same place.
Aman: Indeed. I mean, you all have given us 15 clips in the past five minutes, that we can put on our social media, to share out with the public. I wanna niche down again, no pun intended, but let's niche down into the thinking, to get the bigger picture. The future of public health and health tech, for people that are students and people in the world that are interested in the health and tech field, what are the next, I know Elizabeth, you mentioned predictive AI, but what are some other things that are happening in the next two to five years, perhaps maybe even the next decade, that people might look and to find their niche in health and tech? There might be engineers, there might be other tech people and there might be simply folks that have this love for the public health landscape. What are some areas that we can be looking at that people might usually not know about?
Elizabeth: So one thing is that we are advancing some of the sustainable development goals, for, I don't wanna say a date, but 2030. And we're trying to use digital to advance these goals. And the sustainable development goals are across all different areas. And what you find is that health is everything. So, like for our project for instance, we go to a transportation department, hypothetically, they say, well, and Jose even spoke about this before as well, “transportation, we don't have health data,” but you do, their data is so informative and can impact the health of an individual. And so while you might not believe that nutrition or nutrition data or climate data, air pollution data, can impact health, I think you can carve out and connect, if you're interested in health and you're an engineer and you wanna be more connected into this space, there's a huge opportunity to understand what are the interconnections and actually be a leader in that. Urban planning is a core competency that we're looking for, and there are incredible institutions like Microsoft, Accenture, even investors like Global Fund, that are interested in how to connect technology, innovation in this way. And so there are a lot of different avenues and I would say three really cool projects that we're working on, that no one knows the answer to, kind of like AI for Healthy Cities, sandboxes, test environments for innovation, before it goes into public health systems. So how can you test out innovations or health tech innovations using government data in a secure environment? Kind of like a blockchain for technology. Interoperability, so data standards, data infrastructure is really important. Digital twins is really cool, really excited about that. Personally, they're currently trying to operationalize a digital twin for Seoul in South Korea, and I'm looking forward to seeing how that goes, so we can learn from them, Jose. Sorry, a digital twin is basically a copy of a city environment, so in a virtual environment. So if you have an application that runs on your computer, you could simulate changes to that environment, because it's an exact copy of the infrastructure. So they've done it so far for different airports, like, I think it's Hong Kong International, they're working on it for London Heathrow, Amsterdam. So they've done it for airports, to help planning for airplanes, arrivals and departures and now they're trying to extrapolate that for a city and I'm really excited to see the future of that.
Aman: Wow! Sounds like an enhanced version of Google Maps, that we can't even imagine.
Elizabet: I hope so!
Jose: So I don't know that I can add much more to what Elizabeth has said there, except saying that anything connected to AI, is sort of like, what many people are looking at. One thing that I wanted to say though, was that what we try to do in the department at NYU, is basically provide a set of courses. We do it very consciously, a set of courses that will expose you to many of the things Elizabeth mentioned. So for example, we have a course that we're working on, on how to use AI. I was having a conversation with someone about how AI made you more productive at work. So, we were like, we need a course in that. Like, what are the kind of tools out there, that will make you more productive at work, connected to AI? So, students can get exposed to that. Microsoft is also helping us with a course and we have consulting companies, that help us, where their partners teach courses for us, that are about how do you analyze, for example, health insurance claims data. Getting really deep into, like the whole semester, of a topic that is really only for a few, that are really interested in that. But every semester we get students in those courses and then they come back to me and tell me, "Hey, I got a job doing this and I'm doing something!" 'Cause you realize you learn something later in life class, when you start applying it. But that happens, you know?
Aman: Yeah. Let's pose this final question, because I'm starting to get a sense of the theme now, which I don't often get on this podcast. I have a lot of guests that come, that are researchers. They work in the field and it's amazing the work they're doing. It's very hands-on. But I'm assuming a lot about the collaboration aspect of academia and business, that's there and this is an excellent example of that collaboration, the two worlds colliding together. What makes it so important to pair research with action in this? Can you give me some examples of this? Because there is a lot of students probably thinking in a direction, we've heard of a few cool stories. I'd love to hear some more about where students can learn from this perhaps, something that they can take forward.
Jose: There's something to be learned from each other. And a very concrete example is this project. Microsoft has access to scientists and to ways of structuring data and also store data, that we wouldn't have, if we were just NYU and School of Global Public Health, that's one. And then there's more subtle things, connected to like, if you work with a private sector or with a foundation, for example, or with government, you learn about how they work at different speeds and their priorities and how we push each other in different ways. Academia, on the other hand, I can say it's very good at getting deep into certain topics. That's just the nature of education and how it works. So that has value of course for them too, because, you can put professors and a bunch of students to work on a topic and say, “Here's what's known on this topic – cardiovascular disease, social determinants of health – up to today and we have the tools to do that.” So it goes both ways basically and that's enriching for both. At least that's the perception that I've seen.
Elizabeth: So I think, right now we're talking about a very, I would say also privileged way of thinking in this non-profit way, right? Which is the foundation. I can give an example. So I used to work within the chief medical office, which is a part of the development organization of the company, which is technically not commercial, but it's still embedded into the business. And there were opportunities, we like to move fast as an industry and how can we put checks on that speed and the efficiencies that that will help enable us. Oftentimes the work is being done to check ourselves, but how can we demonstrate that and be transparent about it, to outside authorities, to build trust, to really build that trust? And so, one of my mentors from the company, who is just an incredible individual, started this groundbreaking partnership with actually NYU, around a bioethical advisory committee. So there's outside experts, being consulted on ethical issues related to drug development and this has never been done before, because it is difficult to see things from different lenses, to advance. But the company felt at that time, that it was best to go about it in this way, because that trust is so important. And if we can build trust as a partner, as an organization, you can have these more open conversations when a partner isn't meeting expectations or there's different incentives at play within the organizations. You can talk about it, which makes things a lot more amicable and a lot easier to work through, if you know where the partner is coming from and you give an opportunity to clarify, or to really what I would say is, agency, to the organization to make a decision whether they wanna continue or not. And again, relationships, right? Giving agency to the partner. So I think one of the things that I would want, as a takeaway from partnerships in general, is one of my close friends actually, and mentor, has passed away recently and we found in his wallet, a little blurb, that he wrote to remember about Jack Layton, who was I think a prime minister of Canada, his last words, which was that, "Love is better than anger, hope is better than than fear, optimism is better than despair. If we can be loving, hopeful, and optimistic, then we can change the world." And I truly believe, that if you can create this culture of kindness, of innovation, of openness, that you have a lot more ability to make an impact, or the positive impact, that you want to see. So I would encourage, you can easily be cynical, in this environment and hearing the news and hearing the challenges. You can easily say, "Hey, this isn't working, we're not going to continue." But, I think being optimistic and even maybe a little idealistic actually, I think, will improve and make a better world.
Aman: Love, hope and optimism is what you two have given this podcast. So I really thank you all for your insights. We have learned about deep diving into one thing, committing to it, the power of collaborations and what the possibilities are, in the future of health and tech. You have given all of us the gift and the listeners a gift as well. We hope you really enjoyed this podcast. These are two incredibly busy and high achieved individuals, it's always a pleasure to hear these kinds of insights and aspects of the future, which we usually don't hear of. So I thank you both for all the insights you have given the listeners today and for your time.
Jose: Thank you man. And Elizabeth, it's been a pleasure to share the space with you, over the last few minutes. So, thanks.
Elizabeth: Thank you.
Aman: Absolutely. I mean, thank you both for tuning in. Thank you everyone for tuning into the podcast and folks, we'll see you in the next episode.