Big Data Will Give Physicians Superpowers – Dr. Marc Triola, NYU Langone Health

Shiv Gaglani: Hi, I'm Shiv Gaglani. Today on Raise the Line, I'm really happy to be joined by Dr. Marc Triola who is the associate dean for Educational Informatics at NYU Langone Health, where he's also the founding director of The Institute for Innovations in Medical Education. Dr. Triola's research focuses on the disruptive effects of the present revolution in medical education driven by technological advances, big data, and learning analytics. He has worked to create a continuously learning medical education system that includes new ways to integrate electronic data into educational research. 

As a side note, it's been almost a decade since I first met Dr. Triola at TEDMED 2012, where he was demoing the BioDigital Human, which is a fantastic platform. Listeners of this podcast will know we had Frank Scully from BioDigital on a couple of months back. We also owe a lot to Dr. Triola for setting up this Macy Foundation Conference back in 2015 where our chief medical officer Rishi Desai and I started talking about the concept of him joining Osmosis because, at the time, he was at the Khan Academy. Dr. Triola, thanks so much for taking the time to be with us today. 

Dr. Marc Triola: Thank you so much for having me. I really look forward to our conversation. 

Shiv Gaglani: I know a lot about your background and how you got into medicine and the fact that you still encouraged me to go back and finish med school, which is very likely. Can you tell our audience a bit about your background and what got you interested in a career in medicine and then medical education? 

Dr. Marc Triola: Sure. I grew up in upstate New York. I'm the child of two teachers, always a strong presence of education in my family. When I was a kid, I really wanted to be a marine biologist, actually, very focused on oceans and whales and ended up going to Johns Hopkins, which had a strong marine biology program. As you can imagine, people who are interested in biology who go into Johns Hopkins tend to shift to a little bit more towards the human focus. I ended up becoming pre-med, came to NYU in 1994, and really came to a place that I found fascinating and exciting and stayed in NYU for medical school. I was very interested in Internal Medicine and Hospital-Based medicine. Stayed at NYU for my Medicine residency. I did an extra year as a chief resident. 

During both medical school and residency -- we can talk about this more -- I really saw that information technology, computing, analytics, data, were going to change everything, maybe for the worse, maybe for the better, but they were going to change everything. I ended up doing an additional 2-year Medical Informatics Fellowship here in New York City. I have been back at NYU ever since as a faculty member working with some very interesting people to try to transform medical education and really close the gap between the changes in our health care delivery system and the structure and function of our medical school. 

Shiv Gaglani: That's something that's always impressed me ever since we've started interacting was you not only have very interesting innovations and projects going on -- I remember the pathology or histology viewer that you and your team created adapting it from Google Maps API -- but you also work with very interesting people. One of our first articles published at Osmosis I co-wrote with Jason Theobald, a student of yours and now a good friend of mine who also went into emergency medicine. I actually just saw him back in January. It's great that you've been able to surround yourself with that. Can you tell us a bit about how the IIME came to fruition because I feel like that's, obviously, your incubator of how you have all these great projects and people around you?

Dr. Marc Triola: Sure. The Institute for Innovation in Medical Education is the institute here at NYU Langone Health and at NYU Grossman School of Medicine. It is seven years old and it really stands on the shoulders of giants. My mentor/advisor was man named Marty Nachbar who was a physician here. Believe it or not, he started the original grandparent of our group in 1987. He truly foresaw a future where computers were going to change the way we teach students, train residents, do research, and deliver care. He developed a group called The Hippocrates Project at NYU School of Medicine in the late 80s, well before the internet, and grew that group into, essentially, a software startup within our medical school. It was really quite remarkable. 

That's one of the things that really appealed to me about coming to medical school here...it was to get to work with him. I ended up working with him over the summers and working in his laboratory. He ended up really being my mentor for over 20 years until he passed away just a few years ago. His work and the group and the team that he created -- which I was privileged and fortunate to take over -- started out as this sort of software startup model where we were trying to change how we do medical education, bringing in data from our electronic health record, using new software, new applications, and new ways of doing things.

Seven or eight years ago, we said, "Hey, if we each think about the structure of this at NYU, and we think about it being an institute that could really look across the whole vertical of education -- from before our students get to us, to the application process, all through medical school, residency match through residency and training, and then into practice -- that's what medical education is." It's not these artificial silos of med school, residency, practice, et cetera. If we could create this institute and think about that whole vertical as a continuum for each of our learners, develop new approaches that apply across it all, and really develop our faculty to teach in this world, we could change things. That's where we've been for the past seven years and that's what we've been swimming in.

Shiv Gaglani: That's awesome, again, really impressive. Can you tell us a bit about the size and scope of the Institute as well as some of the projects that you're working on now and what you’re most excited about?

Dr. Marc Triola: Sure. It's a large group considering that we're focused on education innovation. We have 28 faculty and staff who are within the Institute really focused, as I said, on that whole vertical across medical education transformation at NYU Langone Health. The kinds of things that this institute works on are to use data and analytics to transform what we teach and how we teach to maximize the success of our students and our residents and faculty to create this continuously learning healthcare system. It's something that we talked about, as you mentioned, in 2015 at that Macy Foundation conference.

How do we create a system that's constantly measuring itself and constantly trying to improve and knowing that we improve by continuous quality improvement loops? Very importantly, by trying to close the gap between what we teach our med students and our residents and what our hospitals and healthcare systems are doing, and the changing world of our hospitals and healthcare systems. The increasing pace of change -- new technologies, digital health, telehealth and telemedicine, all of the changes this past year of COVID -- those should be a clarion call to every medical school that we need to think about change. We need to keep the core of what we're doing and how we train students and House Staff to be competent physicians, but definitely think about how we are also going to adapt to this rapidly evolving future. 

Shiv Gaglani: Absolutely. One reason we even started talking in the first place was I think after TEDMED, we had published a paper on what medical education can learn from Facebook and Netflix. It addressed what type of big data analytics concepts could be borrowed from those tech companies -- which obviously are no longer in favor because of the social impact that they've had since then -- and apply that to medical education?

One of the most controversial aspects around AI or informatics in medicine, is what's the role of the clinician moving forward? You're more familiar than me about the debates that go on around that. We had Eric Topol on the podcast a couple of months ago as well. He and people like Vinod Khosla have caused a lot of physicians to worry about what their role is going forward. Like, is it going to be memorization and regurgitation anymore? Where do you stand as far as the positives and negatives on the rise of big data analytics both in medicine and healthcare, as well as medical education?

Dr. Marc Triola: I am an optimist. My view is that all of this is not going to diminish or replace physicians. Quite the opposite. The use of tools is going to give each physician superpowers. Most of us think we have them now, but in reality we're limited by our ability to manage lots of information, our cognitive biases, and other things. Right now, we're really good at collecting a lot of digital data about ourselves, about our patients, our communities, our country as a whole. We're not very good at using those data to make decisions, to transform the way we do things, and to transform care. What we need is a new member of the health care team. He's going to help us do that. It's not going to be a human. It's going to be an AI member of the health care team and their job is to give us superpowers. 

Those superpowers are going to be the ability to see patterns, to know what to ignore and know what to look at, to be able to make the right decision for the right patient in front of you. All things that we want to do, all things that many of us think we can do, but in reality, we really can't because of human limitations. These are things that the AI person on our healthcare team is going to help us do. I absolutely do think we should move away from rote memorization and regurgitation. That's what AI is great at. That's what we have definite human limits around. But all of the aspects of healthcare that are unique, that are core to our identity -- the intimacy of that relationship between a physician a patient -- not only should that not be sacrificed in this world, this should never become transactional. But these superpowers should hopefully move some of the rote stuff off of our plate freeing up more time for us to spend with the patients, making our decision more efficient, our ability to review data more efficient. 

I am very optimistic about this. What's really interesting about this future though is that in healthcare, for the most part, a lot of these advances have been physician-facing, hospital-facing, clinic-facing. This brave new world is not limited to just the Ivory Tower in us. Patients, themselves, will have access to all of these tools, too. This is fantastic and exciting. This is some of the democratization that Topol and others have talked about. I think it's absolutely the right thing to do, but managing this world of truly activated and educated patients -- we're not there yet, we've got a long way to go, but we're going to get there -- is going to pose a new challenge for physicians as well. 

I think what we as medical schools need to think about is when our graduates are in the peak of the practice, 10 or 15 years from now, what's that world going to look like? What can we do now to begin equipping them for a world that has AI as a core part of it? We've already got almost 30 FDA-approved AI tools in diagnostics out there in use. 10 years from now, that number will probably be 300 plus. How to know when to use them appropriately? When to deviate from them? When they should increase the space between our data and ourselves, and when they should bring us much closer to our patients? I think that's a big frontier in the future for what we need to teach students to be prepared for.

Shiv Gaglani: That's really well-articulated. We normally ask this question at the end of the interview, but since you're talking about what the next 10 to 15 years may hold, what advice do you give your students -- people like Jason who you've mentored over the years -- about approaching their careers in medicine today?

Dr. Marc Triola: Well, first of all, it's wonderful to see that more and more students are applying to medical school than ever before, especially having gone through a year where healthcare and health systems have been the primary things that have been the pillar of keeping us going. Eventually, rational scientific thought and scientific decision is really ruling the day. It's just thrilling to me that has inspired so many young people to want to take this route and take this path. 

I would say that the emergence of technology, digital health, telehealth, telemedicine, data, and analytics, AI machine, or anything, are opening up much more flexibility and opportunity in people's future practices and their future lives. One thing I would say to people is telehealth and telemedicine may not have been even mentioned during many people's medical school and residency training, but during COVID it was maybe 50% or 100% of practice. It won't be that much in the future, but it'll still be a big component for people. That's a seismic shift in the way that you're interacting with your patient, the way that you're spending your day. 

I think that the flexibility -- the diversity in the eclectic nature of what digital startups, telehealth and telemedicine, new models of care will pose -- means that young people in particular are going to have many more options within the specialties and fields to pursue these. I would say that, increasingly, healthcare for better or for worse, will be delivered through and interpreted through IT systems, electronic health records. Time and motion studies now show us that interns spent 55% of their time doing information management through the EHR, and 13% of their time at the bedside.

Shiv Gaglani: Wow.

Dr. Marc Triola: That doesn't sound right, but maybe that is a great way to deliver care because the EHR is such an important mediator and can help us to make good decisions. Regardless, this information management is going to be the single largest task that anyone in medical school or residency right now, will be doing. Navigating that with fluidity and skill; understanding how to harness these systems to the betterment of your workflow in the care that you deliver and for the experience of your patients...those are going to be really key. The software will get better, the experience will get better, and the tools will give them superpowers. The key will be how well can people adapt to that and redefine their professional identity to get maximal use of all of those things. 

Shiv Gaglani: Yeah. That's, again, really well said with regard to flexibility. Career paths are very winding as both you and I have experienced. Being able to have a prepared mind and take those chances that could lead to great innovations is important. We're really proud to be partners with NYU for a number of reasons. NYU's been at the forefront of a lot of innovations. Your work with the American Medical Association on accelerating change in medical education – the Three-year Medical School Consortium, is one example.

One other thing I'll mention is through the gift of Ken Langone and others affiliated with NYU, you were, I think, one of the first if not the first major medical school to provide free tuition. This is before COVID. Now I feel like the public would very much support whatever we could do to make sure our future healthcare professionals aren't saddled with $200,000 of debt when they graduate. Can you talk about some of those innovations? What are you most excited about as far as the long-term changes that COVID will bring to medical education, and is there anything else that really excites you that NYU's doing outside of your Institute or even within?

Dr. Marc Triola: Well, first of all, I think this is truly an inflection point, but let me say clearly, this was too high a price to have paid to get these changes. This was a national disaster and it certainly highlighted a lot of gap areas, so I certainly wish we could have gotten so much of this disruptive change some other way. That being said, what an amazing time for medical education. I've said often that COVID is the antidote to Flexner. This is the first time since 1910 where things that were immovable objects or unstoppable forces have changed. For example, Step 1 moving to pass-fail; the elimination of Step 2 Clinical Skills as an exam; the virtual interviews for residencies in medical schools which happened with ubiquity this past year. All of those things were really unchanged for decades and changed overnight, just like telehealth and telemedicine changed overnight.

I think that COVID has really forced a lot of people -- many of whom had a lot of momentum and weren't interested in change and many people were really waiting for this opportunity for change -- to say, "What can we do differently and what should we do differently?" Lots of exciting changes happened. Every medical school reinvented itself in one way or another. Every residency program reconfigured itself. Systems thinking became commonplace as people tried to figure out how to adapt and how to manage change on a day-to-day basis. It highlighted the need for teaching and training around informatics, epidemiology, health system science, and many other things.

I think what's next, and what I'm most excited about, is we will be post-pandemic, we will have learned all these lessons, and now we'll have the luxury of choosing what do we want to go back to? What do we want to keep from the rich and storied past of medical education in the United States? And where did we throw out the window during COVID and switch to virtual, hybrid, new? Think about what this post-pandemic curriculum looks like...what post-pandemic residency training looks like? 

You mentioned the three-year pathway which really is a proxy for multiple pathways here at NYU. The concept that we can begin to comfortably move away from one-size-fits-all education, which was largely the Flexnerian model -- if every physician gets X and Y training then we can reassure the American public that they all have a minimal level of competence -- well, now we have much better ways of measuring that competence. We're not ever going to sacrifice that, but we can use that breathing room to provide different pathways for our students, some of which are accelerated, some of which are decelerated because they want to learn more, get dual degrees, do research, et cetera.

The cost of medical education is a huge burden to families. Unfortunately and tragically sometimes, it influences people's specialty choices because they are thinking about revenue and paying back loans and debt. The move here to go to free tuition -- which as you said really did take a tremendous philanthropic effort so it's not likely to be replicated across all 160 medical schools in the country -- is one that we are quite proud of because it really does uncouple the resources for paying for medical school and going to medical school and making your choices about the future. 

I would say, being the director of an institute that focuses in particular on informatics and analytics and technology in medical education, I am excited that this past year has led to a huge increase in the use of online and computer instructive learning. Most of it was replicating classroom instruction and small groups online, but still, we have an army of faculty and learners who are comfortable in this. We're doing this on Zoom, right now. Everybody in the country is comfortable in this. 

What does that mean for medical school, and to a lesser extent, residency training as well? We don't need to be in the same place at the same time. I think that this disruptive change is leading to open-mindedness. I'm really excited to see all the experiments across the country of what it will look like once we have this luxury of choice where we're not forced into a situation by quarantine or pandemic, but we can look back and say, "That really worked maybe better than the way we did it before. Let's keep that or, no, we need to bring in the students. We need to have them together. We need to develop a community, which was hard to do for many medical schools over the past year." I'm very excited about that. 

Shiv Gaglani: Totally. I think this would be, hopefully, a very resilient class of learners and faculty as well, not just in health professional programs, but colleges, high schools, middle schools, and elementary schools, and people who just became more flexible. Hopefully, that flexibility and 'roll with the punches' attitude will continue well after COVID has been contained. I know we're coming up on time, so what have I not asked you that you would love our audience to know about?

Dr. Marc Triola: Well, one thing -- this is a shameless plug for you at Osmosis -- but one project that we have worked on with you guys, which I think is a compelling one, has been the use of electronic health record data to trigger educational resources for medical students. I think that this is of tremendous value. There is so much in the electronic health record about what our students and our residents are doing -- what patients they're caring for, procedures, diagnosis, conditions, treatments, diagnostic tests -- it's all there. Rarely are we able to tap into that, and you use that. It's a gold mine that could inform in a very adaptive way what our students are seeing and not seeing and, therefore, guide their education. 

A project that we worked on with Osmosis has been to create a system where each night our systems within our firewall -- we don't share this data out to you at Osmosis, we look in the Electronic Health Record at patients our students admitted the day before -- look at what were those diagnoses? We've mapped those ICD-10 codes -- this is just for Internal Medicine at present -- to Internal Medicine competencies which then are mapped to Osmosis videos. Each morning, the day after they admit that patient, our students get a push notification from the Osmosis app or an email from Osmosis saying, "Hey, you just admitted a patient yesterday with congestive heart failure. Here is the clinical reasoning video on CHF Diagnostics." Or if they've already seen that one, "Here's the clinical reasoning video on CHF Therapeutics."

I think that this is an important example for a couple of reasons. One, certainly, in the context of COVID, medical schools are going beyond their borders for content. They're looking at companies like yours and many others, and articles, and journals and the MedEd portal from the AAMC. It's really becoming this big ecosystem of content. And, two, we should leverage this environmental and context data of our learners in the electronic health record -- in the teams that they're on, in the concepts that they're learning about in their basic science courses -- to automatically trigger content assessments. No human intervention needed. This, I think, has great potential to create an adaptive learning system.

It's been a fun project. I think at this point, almost 300 video recommendations have been sent to our students based solely on the admissions that they're recording in the EHR on the diagnosis of their patients with no human intervention. We could think of a million other ways to do this -- whether it's triggering assessments and evaluations, procedural observations, feedback -- all sorts of other types of instruction and education. This is one thing that I'm really excited about in the future. It's great that companies like yours, and we need others to follow suit, are opening up not only their content but creating these integrations that schools like ours can take advantage of to make this really a true ecosystem. 

Shiv Gaglani: Well, thank you for sharing that. It's definitely something that Ryan and I, have been talking about since we started Osmosis: could we integrate a recommendation system for professionals, for students, and even for consumers and patients themselves of content that could help them manage their care or provide better care? We found the right partner in you all because, again, you have 28 full-time, faculty at the Institute, and your team's ability to turn around this project so quickly was…I hope more medical schools can follow suit with that because it's been really an incredible work experience. With that, Dr. Triola, I'd really like to thank you, not only for taking the time to be with us on the Raise the Line podcast, but more importantly, for the work that you've been doing for a while now to push the future of medical education forward and, obviously, the patient care that you’ve provided. Thanks so much for taking the time.

Dr. Marc Triola: Thank you, Shiv. It's great to talk to you.

Shiv Gaglani: With that, I'm Shiv Gaglani. Thank you to our audience for checking out today's show. Remember to do your part to flatten the curve and Raise the Line. We're all in this together. Take care.