Speaker 1: Good afternoon, as I said, my name is Ross Frommer, Vice President for Government and Community Affairs at Columbia University Irving Medical Center. Proud to be joined by two colleagues. First is April Smith-Herak, Ph.D. She is the Regional Health Administrator for Region 2, which is New Jersey, New York, Puerto Rico, and the United States Virgin Islands, for the Department of Health and Human Services. As the RHA for the Office of the Assistant Secretary, she is responsible for public health leadership across the region and assists in policy delivery in several areas, including emerging health technology and innovation, women's health, health equity, opioids, COVID-19, and chronic and infectious disease prevention. Dr. Smith-Herak holds a Ph.D. in psychology from Yale University, where she conducted research with the Rudd and Pace Centers, focusing on social and health psychology. Also joined by Commander Michael Banyas, an old friend and colleague, he's a senior officer in the United States Public Health Service and a health specialist at the National Institute of Minority Health Disparities. The mission of his institute is to lead the nation's scientific research to improve minority health, reduce health disparities. In his role, he's Program Director for the NIMHD's SPIR-STDR program and oversees an award budget of almost $12.5 million, which is awarded over 37 recipients. Commander Banyas specializes in underserved health care and public health systems with a focus on implementation science. Previously, he served as a public health analyst at the NIH All of Us Research Program, where he led the Federally Qualified Health Center pilot project and co-led the Tribal Engagement Strategic as well as Operational Process Improvement. Additionally, he served as a fellow on the United States Health and Education and Labor and Pensions Committee, on the Health Policy Subcommittee, and has worked at three academic medical centers. He has a Bachelor of Arts from the University of Vermont, a Master's in Public Administration, Health Management and Policy from New York University's Wagner School of Public Health, excuse me, Public Service. He's a graduate in Health Informatics from Columbia University, excuse me, he's done graduate work in Health Informatics from Columbia University, and a Master's of Arts at the United States Naval War College. Mike, April, welcome. Thank you for joining us here today. We don't have a lot of time, so I just want to get started. April, Mike, could you just talk about the role that HHS as a whole, and NIH specifically, plays in this area of addressing health disparities and how technology is coming to play a part in this?

Speaker 2: Sure. So, HHS, as many of you know, is a very broad agency, department, and it's comprised of several of what we call operating divisions, but most people refer to as agencies. And that includes CMS, FDA, CDC, and of course, HRSA and NIH, which is where Mike is from. So we have this broad health mission, both public health and health delivery, across the U.S. And so many of our components interact in different ways with the emerging health technology. So FDA is obviously responsible for safety and regulatory clearance of new devices and drugs. NIH is spurring emerging research and working on investments in research directed at health. And we have, of course, CMS, who is the largest payer of health care in the United States. So we have very, very broad responsibilities. And in my job, I work in the assistant secretary for health's office. So I work in the office of the secretary, which works as an overarching body over all of the different operating divisions or agencies we see. So my job is really to take a look at the ecosystem as a whole and see where we have responsibilities, where we can help to encourage and spur innovation and work within that health and emerging health technology field in a way that is government and responsible.

Speaker 3: So at NIH, we are the world's largest funder of biomedical research. We have 27 institutes and centers, ranging from our clinical research hospital on campus to the National Cancer Institute to institutes such as mine, which is the Minority Health and Health Disparities Institute, which, unlike cancer or heart, lung, and blood or neurology, those institutes were actually disease agnostic, which is fun for me as someone who wants to work in a lot of different disease areas because I can focus on how to work with my colleagues on addressing health disparities and health equity issues that are affecting underserved populations. So NIH has eight defined health disparities populations. Recently we designated people with disabilities as one. We've also got blacks, African Americans, LGBTQ, low socioeconomic status, rural, Latinos, Asian Pacific Islanders, tribal people, and usually I'm counting this off on my fingers, but my boss told me to stop doing that, so I might have missed one or two.

Speaker 1: You seem to be stuck at one, right?

Speaker 3: Yeah. I started doing it. My boss saw me on a video, and he was just like, don't count on your fingers. I'm in a very unique space within NIH because I run one of the funds for the SBIR program, which is the Small Business Innovation Program, and what we do is, unlike a lot of my other colleagues, which are studying, they call it basic bench science, which is anything but basic biology and science once you actually start getting into a conversation with them, but for the most part they're trying to unlock diseases and the function of the human body. I'm one of the offices that oversees $1.3 billion that's across the NIH that invests in small business and entrepreneurs to come up with commercialized products to carry out the NIH mission, whether it's research or carrying out public health. So most of the companies that I invest in, unlike my colleagues at Cancer, which invest in therapeutics and biosimilars and drugs, I mostly invest in digital health services to help close the health disparities and health equity gap, certain devices, but what we're trying to do, what my fund is specifically trying to do in my program is give out non-dilutive funding, which means that we're not taking any equity, to help spur innovation and growth in sectors that are trying to address people that aren't getting services and close that health equity gap.

Speaker 1: So we talk a lot about health disparities. What areas do you see that these disparities are the greatest? Where is the need the most, both in terms of the population, and I know, Mike, you said you're not disease-specific, but I'm going to ask you a disease-specific question, where do you see these disparities evidence themselves the most?

Speaker 3: Genetic diseases, a lot of the diseases that are treated in such as federally qualified health centers like diabetes, stroke, hypertension, eye issues, a lot of diseases that just over time impair people's lives that might not, in which treatments and education just might not be culturally competently reaching them, they might have access issues, they might have economic issues, or there might be a knowledge barrier as well. So chronic diseases, I know a lot of researchers want to go for that miracle drug, but chronic diseases are where, whether it's rural populations, African American populations, or Asian populations, those are the disease areas that have the greatest health compromise in those individuals.

Speaker 1: You talk about the various populations, are there certain populations where there's disparities can be greater, or the disparity can be greater than others, or?

Speaker 3: Yeah, such as diabetes is huge in Asian populations, COPD is big in tribal and rural populations, and those are just some examples. It all varies by, sometimes genetics can play a factor, lifestyle, or social determinants of health, which is how someone's health is impacted based on their environment.

Speaker 1: So there's been a lot of talk about digital health, and obviously over the last three to four years, we've all learned a lot more about digital health than we ever thought we would. How can digital health help us, the government, the country, try to close some of these gaps?

Speaker 2: I think digital health has an opportunity to increase some of the access issues in a positive direction. So I think some of the lack of access to health care, especially for populations in rural areas where you just don't have that same volume of care, I think making digital solutions available, the advent of things like hospital at home and remote care, easier access to medication-assisted treatment for substance use issues, I think that all of those emerging technologies that have really come into fruition in the last several years have given us an opportunity and also a great danger. If we implement them properly, what it can do is it can provide additional access and better care for everyone so that you do eliminate some of those disparities. Implemented unevenly, I think what we see is an opportunity to exacerbate these disparities and make them considerably worse. So as a governmental and public health entity, one of the things that we're always looking at is how do we make that first scenario the one that we're trending towards and not, you know, developing these emerging technologies in a vacuum that just continues to increase the disparities in access to care and health outcomes.

Speaker 3: I think probably, you know, I know the pandemic was especially terrible for New York City, but I think one of the interesting things or one of the few positive things that came out of it was the expansion of telehealth. I think you're seeing like right now in Congress, they're trying to figure out how to do reimbursement, keep permanent reimbursement for telehealth services. And you see that with the increase in access and probably most strikingly in the behavioral health aspect. Like it's, you know, access to behavioral health providers has always been a huge issue. And I think you're seeing, you know, I think digital health is a huge example about how that has helped revolutionize providing mental health access to people. In addition, diagnostics, like digital health is going to have a huge area, continuing huge area in diagnostics such as diagnosing skin cancers or eye diseases. And it's going to be really interesting to see what other tools are going to come out that are going to help with the expansion of primary care and helping physicians and the care team actually track symptoms and monitor patients from that perspective.

Speaker 1: You know, it's interesting in the spring of 2020 and the months beyond, as an advocate, as a policy expert, I spent a lot of time doing work on virtual health issues, trying to remove those barriers, trying to make it more accessible, not only for technology issue, but from a regulatory point of view as well. And still some issues we need to deal with both on the federal level and also on the state level. We only have a couple minutes left, but I want to talk about artificial intelligence, AI. It's a buzzword. It's used a lot today. How can AI be used, especially with managing large amounts of data, to sort of put that data into good use, if you will?

Speaker 3: So this is an area, like, so recently there was an executive order that was put out on AI. And I just want to clarify what an executive order is and how it differs from regulations. Regulations are spurred through legislation and the creation of a law. An executive order comes through, you know, the president's office, and it impacts the way that, like, you know, that grants are formed and programs operate. But it's not law. So with the AI stuff, recently there was an executive order focusing on privacy issues and access. And we're still trying to digest it. Actually, April and I are talking to the HHS AIs on Friday. So if I'm kind of trying to choose my words carefully, it's because we're still trying to figure out our boundaries with it. But I think AI has an enormous potential. But as someone who gets a lot of, who talks to a lot of AI companies or digital health companies that are applying to NIH for funding, you know, it's going to be really interesting. It's always very interesting to me to see, one, what type of service they're using. Are they using Amazon? Are they using something homegrown? Are they using Google? Like, really, there's three big players in this area. Amazon, Google, and Microsoft. You know, who are they using for their backend service? Where they're getting their data. That's always a huge question I'm asking. Like, I liken AI to being a really smart person. But if they're not pulling the right books from the library and they're just pulling, like, conspiracy theories, well, then you're going to have bad information about bad outcomes. But, you know, it's going to be very interesting to see how it actually changes clinical trends. You know, the way doctors and nurses and care is delivered to patients and how clinical decisions are made. But also how you're able to take the vast amounts of data from research and actually link it together and put together a coherent picture for greater health outcomes in research innovation.

Speaker 2: I think that there's a kind of a difference between clinical research and AI applications. And the kind of AI applications that are coming from existing data sets, which often come from claims data from insurers and so on. And I think that it's really important to be considering where the sources of that data solution is. Because the thing to understand is that we already exist in a biased system, that our system is already imperfect. So if we do end up training our AI algorithms on biased data, then what we get is a biased algorithm. And those are some of the concerns that I think have come up as we've begun to discuss AI as a government entity. There's a lot of decisions that need to be made. And there are places where, you know, unbiased, really very like scientific data oriented data going into an algorithm is a different conversation than an algorithm that's being fed from a system that's already incorporated bias into that system. So how are we protecting the American people from the potential misinterpretation or misuse of data by an AI system? And that is a little bit of a hairy conversation for us to figure out.

Speaker 3: It's a hairy conversation, and it will be interesting to see if there will be congressional action in upgrading HIPAA to align with what's coming out with AI. There's obviously a lot of privacy laws and being in the health disparity area. You know, there's a huge amount of questions about what information, you know, we're going to have access to and be able to use as far as addressing health disparities and addressing people that are in need, but also maintaining that privacy for people. But a lot of it is going to be where the information is coming from. You know, like I go to the Autumn Conference, which is the big tech transfer conference, and a lot of medical schools are now selling old research data as IP because now it can be used for something. But where is this data coming from and how it's being utilized? Because that's going to, you can make very dangerous care decisions if that data is not evidence-based and it's not valid.

Speaker 1: Well, with that, I think we're going to have to wrap up. I'm getting the eye from the guy on the side of the stage. So I want to thank you both for joining us here today. Thank you very much, and enjoy the rest of the conference.