AMA Update covers a range of health care topics affecting the lives of physicians, residents, medical students and patients. From private practice and health system leaders to scientists and public health officials, hear from the experts in medicine on COVID-19, medical education, advocacy issues, burnout, vaccines and more.
Featured topic and speakers
Featured topic and speakers
Is AI taught in medical school? What is AI integrated education? How will AI impact health care in the future? What are the benefits of health informatics?
Our guests are Kim Lomis, MD, vice president of Medical Education Innovations, and Margaret Lozovatsky, MD, vice president of Digital Health Innovations, at the American Medical Association. AMA Chief Experience Officer Todd Unger hosts.
- The AMA is your powerful ally, focused on addressing the issues important to you, so you can focus on what matters most—patients. We will meet this challenge together. Join us.
- Looking for generative AI in medicine courses? Access AMA EdHub Introduction to Artificial Intelligence in Health Care.
- For more on ethical issues in artificial intelligence, check out AMA Framework for Health Care AI.
- Learn more about our AMA advocacy priorities, including:
- Reforming Medicare payment
- Fighting scope creep
- Fixing prior authorization
- Reducing physician burnout
- Making technology work for physicians
Speakers
- Kim Lomis, MD, vice president, Medical Education Innovations, AMA
- Margaret Lozovatsky, MD, vice president, Digital Health Innovations, AMA
Unger: Hello, and welcome to the AMA Update video and podcast. Well, it seems like everyone's talking about AI all the time, and especially how it relates to school. And medical schools are no exception. Today we're going to talk about how medical schools are incorporating AI into their curriculum.
I'm joined by two guests from the AMA. Dr. Kim Lomis is the vice president of medical education innovations in Chicago. And Dr. Margaret Lozovatsky is the vice president of digital health innovations in Charlotte, North Carolina. I'm Todd Unger, AMA's chief experience officer, also in Chicago. Dr. Lomis, Dr. Lozovatsky, welcome.
Dr. Lomis: Thanks. Hi, great to see everyone.
Dr. Lozovatsky: Thanks for having us.
Unger: It's exciting to talk to both of you. My first question today is for you, Dr. Lozovatsky. AI has been used in medicine for a while now, but it hasn't prompted a major change in medical education until recently. Why the gap?
Dr. Lozovatsky: That is absolutely correct, Todd. AI has been used in health care for a very long time. We've used a lot of predictive modeling and ambient listening, and it's been heavily used in areas like radiology for many years. And yet in 2023, the change was ChatGPT. When generative AI came out, people really started to notice this new technology, and it brought a lot of attention both to health care organizations, to physicians and to technology stakeholders.
And so we are now seeing this amazing capability of the technologies that are out there to address administrative burdens and to help with cognitive burdens that physicians are experiencing. And so as we are seeing more and more investment in this technology, this topic has become integral to our conversation in health care informatics and clinical technologies.
Unger: Well, given the impact that AI could have on medicine, gosh, there's a lot for medical schools to potentially teach students. What do you think are the most important concepts for them to focus on right now?
Dr. Lozovatsky: We like to break it down into five major categories. The first one is foundational knowledge, just basic understanding of what is AI—what does it do, how do these technologies work. Then thinking about critical appraisal. So what's the value of using these tools?
We know that they can reduce administrative burden, as I mentioned, thinking about cognitive overload. How do we incorporate them into decision making? When and where should these technologies be used?
We know that AI enhanced clinical encounters are very important. We also need to think about the technical considerations. So workflow considerations, what are some of the cybersecurity risks? How do we use these technologies appropriately? How do we bring them to the bedside for the clinicians to be able to integrate into clinical care?
And finally, we need to understand the key risks and potential unintended consequences that can occur with these tools. So understanding that this is a process that's going to be continuously changing and improving, these models need to be thought about regularly. We need to consider social and ethical areas such as bias and understand how they can impact clinical care.
And finally, I want to highlight that these concepts are important for all physicians to be learning. That starts from undergraduate to graduate medical education and also in continuous professional development. We need to integrate it into all levels of learning.
Unger: And it does sound like a great framework for how to think about it. Dr. Lomis, I have to believe that on the other end of this, educators are saying another thing that we need to teach medical students on top of everything else. How should medical schools go about incorporating concepts like what Dr. Lozovatsky just went through into their curriculum? And have you seen any approaches that seem to be working especially well?
Dr. Lomis: Thanks, Todd. So what we have seen mostly so far has been elective offerings largely driven by interested students. I think we all know that it's often our students and residents who elevate these emerging topics and advocate for their own training. But it has now become urgent that we as leaders of medical education purposely put this into required curriculum.
And you're right, that can be challenging, especially because many sites feel that they don't have the expertise. We've done one step here at the AMA to help give people a leg up. We've created a series of online modules that are available on the Ed Hub, which is a great start. There's seven modules that are very foundational that can give a site a starting point and then build from there to incorporate this into their own programming.
The good news is that AI touches everything that we already teach. And so if we're strategic about it, once the learners have a little bit of foundation, you can weave this through things that are already happening in the curriculum. And that's an effective mechanism we have seen. When schools are teaching about evidence-based medicine and clinical reasoning, this fits perfectly into that. And honestly, it's unfair to not mention it in this age if you're teaching those topics.
Similarly around interpersonal communication skills, increasingly, patients will be coming with data and having questions about data. And we need to explain, as Dr. Lozovatsky said, how we're using these tools to those patients. So this can be fit in a lot of different places. And one, in particular, I always try to emphasize to our learners is in documentation. We are generating the data sets upon which these tools are going to be trained. And so some of the sloppiness that we have sometimes seen in documentation becomes actually somewhat dangerous, because it can be amplified through these things.
So we can find places to put this in. It doesn't require inserting weeks and weeks of training.
Unger: Well, it's interesting because, on one side, what you're talking about is, of course, what medical schools need to talk about to teach students about AI. But as we know, just from education writ large, they also have to teach—think about teaching students with AI. How have medical schools started to do that, Dr. Lomis?
Dr. Lomis: So our meta team at the AMA super excited about this opportunity to leverage informatics and AI to improve the process of education. We're referring to this as precision education, essentially, knowing enough about an individual based on their activities to target their learning and individualized pathways. We think this will not only make education more effective, but it can impact well-being as well.
So especially at the residency level, we know that there's some dehumanizing if it's not targeted to what you really need. So we've been sponsoring some great projects and funding, some exploratory grants with partners such as NYU and OHSU and many others. And so there's some neat ideas that are coming out. It's possible to take a look artifact of a learner, whether it's a student or a resident, entering a note about an encounter with a patient.
And some of these sites are then using natural language processing to interpret the note, to give the learner credit for key concept, populate that to a dashboard to see what they have and importantly, have not yet experienced, so that you can be more targeted with the clinical exposure. And so you need to go get more experience in this area because you haven't had those concepts yet. These sites are also immediately leveraging that information to suggest learning resources in the moment that are vetted by their program.
So I think there's just tremendous opportunity here, and we will be continuing to support this at the AMA with a future initiative coming in the near future.
Unger: And I really love how you're talking about precision education and really looking to AI to bring that kind of dream into reality with things moving so fast. Obviously, there is a lot of change in a short period of time. Dr. Lozovatsky, how can medical schools keep up with such a rapid pace of change?
Dr. Lozovatsky: Yeah. Clinical technology is moving, like you said, at a very rapid speed. And we know that today it's integral to our ability to care for patients. It's in everything that we do. And so I'd like to share a few guiding principles that we think about as medical schools are developing these curricula.
The first one is all complex concepts can be broken down. And so just like all of the other topics within clinical technology that we've been thinking about for a long time, this is a topic that can be broken down into the areas that we discussed, thinking about the basic facts and then how do we integrate it into clinical care.
Dr. Lomis mentioned precision education and how we can use applied learning. This is very critical to our ability to be successful in educating on this topic. I think about the days that I am taking care of patients. When I have students, I use every opportunity where I see some of these models in clinical practice to show them to them, to explain how documentation drives the results that they're seeing there and how they can use these models to be able to drive their clinical decision-making.
I also think it's very important to integrate expertise into our institutions. And we're still learning what that expertise might look like because these technologies are evolving. But we do know that most organizations have clinical informatics departments. This is now a board-certified specialty, and there are people that are trained to understand how to bring these technologies to the bedside. So leveraging those individuals in these conversations, in the curriculum development, is going to be very important.
We need to educate our faculty. It's very important for all of us to be able to communicate this to our learners. So we need to be knowledgeable, and we need to keep up. So that's really the first step in this educational process. And then finally, as you mentioned, because these technologies are moving so quickly, we have to review our programs, competencies and curriculum regularly. It's going to require a commitment of time and energy to make sure that we continue to provide the most up to date information to our learners.
Unger: Dr. Lomis, I want to follow up with you about something Dr. Lozovatsky mentioned, and that is, of course, the medical educators, who, obviously, have a huge opportunity here but need training. How are they feeling about all of the dynamics involved in AI and teaching that?
Dr. Lomis: I appreciate you asking that, Todd, because it can be overwhelming. I've personally found that you have to learn a little about AI to recognize how much you need to learn about AI. And I think that it's easy for educators to pull back and see it as futuristic, not sure that we can fit it in at this time. And I would encourage any educator to just dip your toes in enough to see how radically this will change the practice for our future trainees, our current trainees in their practice in the future. And so we have to invest in and jump in.
The AMA is a member of the National Academy of Science, Engineering, and Medicine Forum on Health Professions Education. And we help to create a discussion paper. And that would be a great reference for a program who was trying to figure out where to even get started. There's a discussion paper on their website around artificial intelligence for health professions educators. And that has, at the end of it, a step by step of where you should look locally for resources to help you engage and jump in.
And then additionally, for those who are early adopters and excited about using AI to help improve the process of education, the AMA has also issued guidance advancing AI in medical education through ethics, evidence and equity to remind us, as we use these tools for educational purposes, we have to be mindful of the same bias, data safety and privacy issues that you would use in health care. And so we do have some resources, but it's an ongoing effort among the educators to learn and ramp up.
Unger: So we have a lot of resources, obviously, from the AMA. Dr. Lozovatsky, for people who have more questions, want to learn more about AI, any recommendations on where they go to learn more?
Dr. Lozovatsky: Yes, absolutely. As Dr. Lomis mentioned, we have spent a lot of time creating some of these materials, and we would love to share them with you. And there's a few areas that we have covered. One of them is advocacy principles. Those focus on both device and non-device AI-enabled technologies. And they also assist in information about deployment of these new technologies into your clinical areas.
We have AI research that has been done over the last year that really covers the landscape of AI in health care. It includes terminology, current and future use cases, and also addresses some of the areas of risk that I know are very top of mind to most of us.
We have, as Dr. Lomis mentioned, the AI learning series modules on our Med Ed section of the website. And those materials are led by Dr. Lomis's team. And we also have additional materials on ethics and CPT and how AI touches those spaces. There is a website on the AMA site that specifically focuses on AI. So please visit that area and take a look. There's many, many materials there.
Unger: Dr. Lozovatsky, Dr. Lomis, thank you so much for joining us today. We're going to include links to the resources that Dr. Lozovatsky just mentioned and Dr. Lomis throughout this episode. But they'll be in the description of this episode. So make sure to look for that information and go find out more.
The AMA is fighting to make technology work for physicians, not the other way around. And that starts, of course, in medical school. So if you want to support work like this, become an AMA member at ama-assn.org/join. That wraps up today's episode and we'll be back soon with another AMA Update. Be sure to subscribe for new episodes and find all our videos and podcasts at ama-assn.org/podcasts. Thanks for joining us today. Please take care.
Disclaimer: The viewpoints expressed in this video are those of the participants and/or do not necessarily reflect the views and policies of the AMA.
Subscribe to AMA Update
Get videos with expert opinions from the AMA on the most important health care topics affecting physicians, residents, medical students and patients—delivered to your inbox.
