[MUSIC PLAYING] OBY: From the campus of Harvard Medical School, this is ThinkResearch, a podcast devoted to the stories behind clinical research. I'm Oby. 

BRENDAN: And I'm Brendan. And we are your hosts. ThinkResearch is brought to you by Harvard Catalyst, Harvard University's Clinical and Translational Science Center. 

OBY: And by NCATS, the National Center for Advancing Translational Sciences. 

BRENDAN: Over the last 70 years, health informatics has spurred the development of many different technologies for tracking, processing, and analyzing patient data. On today's episode, join us as we talk with Dr. Andrew Marshall, Instructor of Emergency Medicine at Brigham and Women's Hospital, and discover how combining software engineering and medicine can lead to more equitable care and better health outcomes. Drew Marshall, thank you for joining us. Welcome to ThinkResearch. 

ANDREW MARSHALL: Hey, Brendan. Really nice to be here. Thanks for having me on the show. 

BRENDAN: So you're an Instructor of Emergency Medicine at Brigham and Women's Hospital. Could you tell us about your choice to pursue emergency medicine? What interested you in that specialty? 

ANDREW MARSHALL: Yeah, so I think when people think of emergency medicine, they think of a jack of all trades. And as I was going through medical school-- and I think a lot of people will have the same story. They just liked a little bit of everything. And I think there's some other specialties that I was interested in. I really liked surgery, but I didn't really like the lifestyle. 

And I really liked OB/GYN because I think they practice in the operating room, they practice in the clinic, they practice on the hospital floor as well. But at the end of the day, I feel like I had a lot of interest outside of medicine as well, and I felt like emergency medicine gave me the best lifestyle, the best schedule, the ability to have a family. And really, I still feel you could pretty much drop me anywhere, in any hospital across the US or any kind of care setting, and I have some useful skills that I can do. 

So I think it's really cool to know a little bit about a lot of specialties. I think one of the things that maybe consultants find frustrating with ER doctors is that we don't know the depth of knowledge that they know, but we know how to call the right people. And I think it's a job that I still find really exciting, even through the pandemic. 

I think it's been something that has given me purpose and something that I still find really enjoyable. I still like going to shifts because I think they are pretty fun, and you never know what you're going to see, and you never know who you're going to meet and who you're going to help. So I think those are all reasons that I love emergency medicine. 

BRENDAN: Before you decided to go to medical school, you interned at NASA and were a software engineer. What inspired you to make the switch to medicine? 

ANDREW MARSHALL: Yeah, so I think the full background is that I was a little unsure about whether I wanted to do medicine. I think there was always a push from my mom to do medicine. When I was growing up, I fell in love with video games. And I was like, how do you make video games? And as a kid, I wasn't really allowed to play video games, and so I would sneak and play my friends' Game Boys at school and that kind of stuff. 

And so that got me into, how do you make these? Which is software. And so instead of studying biology in college, I decided to do computer science. And I really enjoyed computer science. I think there's some beauty and the creativity of it. I think people think it's all math and that kind of stuff. 

But there's some real design and creative thinking that goes into, how am I going to solve this problem? But the whole time I was taking computer science, I took the prerequisites for medical school because I always viewed biology as this kind of-- at least the way it was taught when I was in high school and college is that you memorize this information, and then you go and spit it out on a test. I didn't enjoy that aspect of it. I don't enjoy memorization. I'm more of a big picture person. 

But it wasn't really until my junior year when my dad got sick-- he actually ended up in the hospital with a subdural hematoma and was rushed to emergency surgery-- that I really cemented my decision to go into medicine. Essentially, he had been to the hospital a couple times with really bad headaches after being started on a blood thinner for a sports injury where he had developed a blood clot, and they had basically sent him home twice with really severe headaches. 

And then on the third try, he went to a different hospital, got a repeat image, so I guess they had probably missed the bleeding on a previous image, and then was rushed to emergency surgery. And this was right at the beginning of my summer. So I remember coming home from college. Was getting ready to go to North Carolina to do a program in biophysics, which is probably a little bit more on the research computer science side of things. 

And I remember waking up and him and my mom not being there, which is weird because they'd normally be around, kicking around, making breakfast, that kind of stuff, and then hearing the news in the hospital and rushing down there. And I realized that this was a big miss. If they had caught this earlier, I think they could have stopped blood thinner. Knowing what I know now, I think it could have changed the course of what had to happen. 

But instead, he was rushed to this emergency surgery. It was then that I was like, OK, there's some things in medicine that need to be solved. There's some ways to make things better. And the engineer part of my mind was really intrigued by that. And so I took the MCAT once. 

Didn't do that great on it. Took it again. Applied to med school. I don't think I ever really got the MCAT score that I wanted, but the skills that I brought from computer science and engineering I think were tremendously useful and helped me be successful in medicine when I finally did get it. 

BRENDAN: And just so people know, your dad is doing OK now, right? 

ANDREW MARSHALL: Yeah, he's doing great. Yeah, he's doing really well. I'm actually at home with him right now, and probably after we get off this podcast, we'll go play a little bit of tennis. But very, very lucky to have him around after such a scary thing. 

BRENDAN: That's great to hear. So you mentioned your dovetailing interests in computer science and medicine, and I think that's where your work with informatics brings those two things together. So talk a little bit just briefly about what informatics is and how your engineering brain and your medicine brain fuse and do that work. 

ANDREW MARSHALL: It took me a long time to figure out exactly what I wanted to do. So I was in med school, and I had these skills, and I built things here and there. And I went to residency, and I was like, what should I do? And then I found out about fellowships in informatics. And informatics is basically-- I think there's two different ways of looking at informatics. 

Just bioinformatics, which is how you look at data, which is a child of data analytics, and then there's clinical informatics, which is about how you implement these really cool data tools, AI, clinical decision support, those kind of things into medicine. And so I ended up doing a little bit of both. 

Because I have programming skills to do some data analytics, I thought it would be-- I really wanted to get into AI. And so I basically did a fellowship in clinical informatics and got a master's degree in bioinformatics at HMS. And so I think it's really cool because it takes you out of just the individual practitioner role, and you get to think about problems at a system level. 

So when I talk about clinical decision support, these are the kind of things that I think about helping people like my dad. So there are AI tools that will read radiology reports and suggest things to the radiologists that they might have otherwise missed. And then the way clinical informatics works is you're the person that understands a little bit about the informatics world, so the computer science world, and the person that understands a little bit about medicine. 

And so you're the person that can talk to the radiologist as well as the person designing the system and say, oh, this is the best place to insert this into the workflow to make this happen. And so that's how those two meld. So I view clinical informatics as offering clinical decision support to improve the practice of medicine through synergy. 

So I don't ever think that computers are going to replace what I do or what any other doctor does, but I think there's got to be a synergistic relationship that allows practitioners to practice probably at a higher, safer level than they would have been able to practice on their own, kind of like the support that you get from a car that tells you when you're changing lanes or is able to help you park, parallel park, when you're not a good parallel parker, all these really cool new technologies. And I see that being able to really enhance medicine rather than taking anybody's job, which is a big fear of the field. 

BRENDAN: Right, right. Radiologists would just be replaced by algorithms that read-- 

ANDREW MARSHALL: Right. By IBM Watson, right? 

BRENDAN: Yeah. I'm sure IBM would like to think that they'll replace every radiologist, but-- 

ANDREW MARSHALL: They're not even close, in my opinion. 

BRENDAN: That's good to hear. So one of the projects that you're working on that uses analytics is a screening tool for patients who may be good candidates for home hospital. Tell us a little bit about what home hospital is and why this project's important. 

ANDREW MARSHALL: So first, I want to give a shout out to collaborators because none of these things happen in isolation. And so David Levine and his team are the people that originated home hospital, and I work with Phil Anderson and Yonatan Keschner and Zoe, who is a data analyst on this as well. Home hospital is really important, and it's become more important because of the capacity that our hospitals have been at. 

And this is since the pandemic. And I think, in my mind, I'm just a little unsure about why we've been at such crazy capacities. I think there have been waves of COVID, and that's very explainable, but there have been a lot of other-- I think people are, in general, sicker. I don't know if it's because they stayed away from hospitals or if they're experiencing side effects of COVID in the long term, but we've had hospitals that have stayed full for pretty much the past two years. 

And so our ICU capacity is full. Our inpatient wards are full. We're having waiting rooms that are 20-plus, 40-plus people, which is pretty unheard of in Boston, which traditionally, it's thought of as one of the health care hubs in the world. We've got a lot of really great hospitals. And for them all to be at capacity is a really unusual thing. 

And so home hospital is a program that really thinks about using technology in a creative way to allow patients to get their hospital-level care at home. And so everybody, almost everybody-- I don't want to say everybody because there are people that don't have homes, but almost everybody has a home. And if you're able to provide the care that they need at home, you're able to offload some of that pressure in the hospital. 

Now, a lot of the questions around that are, well, are you going to provide subpar care for them at home? And the answer is no. You just carefully select patients that are going to do well if a doctor and a nurse come and visit them from the hospital and maintain their antibiotics or maintain all that stuff. And so I think that as we look to scale this project up, there's a need for being able to select patients quickly. 

Manual processes, I think everything right now is manual, so if I'm working in the emergency department, somebody may come up to me and basically ask me if I have any patients that they think are good to go to home hospital. And that's subject to bias. So I may look at some patients and be like, yeah, these are good to go, these are not good to go. But we may be missing a significant amount of the patient population. 

So I think a really cool thing is what we're building is an ability to automatically screen for those patients. And so whereas a clinician like myself, somebody on the home hospital team might sit in the morning and manually review all the overnight patients that were admitted to see if any of them could go to home hospital or manually review all the patients in the ED to see if any of them are suitable or, god forbid, manually review everybody that's an inpatient. 

We want to automate that process so that they get a quick answer and say, oh, we just screened 300 patients on the medicine ward in less than a minute. And these are the ones that are your greens that can go to home hospital. These are the ones that are your yellows that maybe could go if we change one or two things. And these are the people that are really too sick, and they should stay in the hospital. 

And so I think that that is going to really allow the home hospital program to scale. And if it gets bigger, I think that's better for everybody because then we can provide appropriate levels of care, decompress the hospital, and overall, just take better care of our patient population. 

BRENDAN: Great. So I want to talk about your work with the Office of IDEAS at Brigham. You're the Social Justice Committee chair for the Office of IDEAS. Can you tell us a little bit about what the office does and what you do in that role? 

ANDREW MARSHALL: Yeah. So the Office is-- so IDEAS stands for Inclusion, Diversity, Equity, and Social justice. And I was lucky when I came on to the Brigham two years ago after finishing fellowship that they were starting this brand new office, and they invited me in to-- I don't want to say be one of the founding members, but be close to the founding. So I was the first social justice chair. Still fill that role right now with DaMarcus Baymon, who became my co-chair in the last couple of months. 

What we do is we look for actionable projects to improve inclusion, diversity, equity, and social justice. And so while the Office of IDEAS is not necessarily research-focused, I think we think more about, what can we do to actually make a tangible difference? I think there's a lot of really cool research projects that end up coming out of the cool work that we're doing. 

BRENDAN: So one of the projects that has come out of this office that you're working on is about stroke education for high school students. Tell us a little bit about that. 

ANDREW MARSHALL: Yeah. Again, you got to give full credit where it's due, and as the chair of the Social Justice Committee, I'm more of a sponsor and a mentor. So for full credit, I've got to give a shout out to Mariama Runcie, who's one of our stellar residents. She's a former teacher, and she assembled this multidisciplinary team from MGH and HMS and I think the business school as well to think about how we could educate patients about stroke. 

And so we decided to target high school students. Since Mariama is a high school teacher, she has a lot of experience working with high school students. And the idea behind this is that you've got to empower patients to make the right decisions about their health care. You don't want anybody to ignore the signs of stroke, god forbid. And you don't want anybody to not be able to communicate with EMS or the ambulance when they come and get them because when you have the signs of the stroke, you need to go to a stroke center. 

And Brigham and Women's Hospital is a stroke center. There's several other stroke centers, MGH, BI across town. And what ends up happening I think sometimes is people, if they don't recognize the signs of a stroke, may be less reluctant to advocate for themselves to go to a stroke center. So I think this is like a social justice issue. It's a health literacy issue. 

And I think that one of the coolest things that we did was that we had the opportunity to go into high schools, even though we did this in the middle of COVID, and offer live teaching, offer these video recordings, and then offer pre and post-tests for high school students and see how they did. And there were some really exciting findings. 

I think one of the funniest things that came out of this was that students that did not receive some sort of formal stroke education would fill out the pre-test, get some questions right, go on the internet, look things up, and then come back and get them wrong on the post-test because there's so much confusing information out there that can trip you up. But the students that had the video recording or had the live teaching, they did really well. 

And I think this is a great pilot. That's how we viewed it last year. This is a great pilot for a program that could be expanded to do in multiple classrooms, community health centers, just educating people about stroke. And it's a great example of almost like a noninferiority trial. Can you do the live teaching? Can you do a video recording? Do they have the same outcome? 

And I think the students received it really well, especially because we worked in a lot of classrooms with Black and Hispanic students. And Mariama is a Black female physician, which was really well-received. So I think they enjoyed learning from somebody that looked like them. So I think that was a really powerful project and one of many really cool projects coming out of the office. 

BRENDAN: Yeah. That's cool. I like what you said about just having somebody who looks like the students in a doctor role. They're learning about stroke education, but they're also learning about career paths, and they're learning something that is maybe less, I don't know, tangible. 

ANDREW MARSHALL: Yeah. A lot of people say, you can't be what you can't see. And so just by being there, I think she definitely made a big change, and I think that that's part of what we want to do through mentorship in the office is just do some cool stuff, help some people. We provided scholarships last year to a bunch of kids, but also just be there and just represent and show that there's a path forward in medicine and whatever else you want to do. 

BRENDAN: So you talked, at the beginning, about the experience you had with your father and his hematoma and thinking the system kind of failed him. And that prompted you to think about equity in medicine. So as you do this work with the Office of IDEAS at Brigham and just your practice, how do you see your work in equity and informatics continuing? What do you see as the future of this work? 

ANDREW MARSHALL: I think it's an interesting problem to solve, one because I think when you think about technology and medicine, you can see a lot of examples where lack of access to technology can deepen health inequities. You can see that lack of health literacy can make things a little bit worse because we have this one size fits all approach, where we assume that everyone can use the tools that we, as English speakers, we as people who are familiar with technology and have had access to technology our whole lives, we assume that they can use those same technology. 

So I have a vision of both making technology more available, more accessible, and using it to close those gaps in health equity. I think we got to start by getting kids, parents, patients access to broadband and devices. I don't want to say that broadband access should be a right, but it's really hard. Imagine if you didn't have broadband access during the pandemic, and you were a kid in school trying to get an education. 

You had to Zoom in to everything. Imagine if you're a patient, and your clinic has gotten shut down in the pandemic, and you had to do telehealth and you don't have broadband access. So you can see that these things are affecting the social determinants of health. They're affecting education, and they're affecting your ability to even see your clinician. 

So we also got to think about using technologies to communicate with patients in the way that they would like to be communicated with. And I think that this involves thinking about language. This involves thinking about tech literacy levels. I think that telehealth can take a lot of different forms. So telehealth can be a phone call, telehealth could be a text message, telehealth can be an email, telehealth can be a very complex app that you navigate to do video chats and stuff like that. 

So I think taking all those things into account, we have to build a better health care system that both makes technologies accessible when we need it for our patients, and coming back to what I said earlier, accessible not only physically having the technology, but accessible in being able to understand what's going on. I look forward to pushing that agenda, to building tools that patients can use in the emergency department and elsewhere. 

For example, I think one of the things we haven't talked about is that I've partnered with E Ink. Dr. Chai and myself, Peter Chai, we built these screens in the emergency department that basically provide information to patients real-time during their visit. And the first iteration of this project was only English-speaking patients with information that is provided in a contextual way. But we saw that satisfaction increased. 

And so thinking about this project going forward, if we could provide patients that are traditionally marginalized, like non-English-speaking patients, patients that don't have the ability to have a cell phone in their hand and access patient gateway while they're in the hospital and look at their information, we're really doing a service to them getting more engaged and taking ownership of their health to move things forward. And we know that when patients are more engaged, they have much better outcomes because they have better understanding, and they're more willing to make the changes they need because they can understand those changes. 

BRENDAN: So using E Ink screens to show patients information about their visit while they're in the hospital. Can you tell us a little bit more about that? 

ANDREW MARSHALL: So we're working on publishing these right now, but we did look at-- in a very small sample size, actually, we looked at how patients reacted to getting their information in real time. And it actually showed, in some categories-- because we used Press Ganey, which is the satisfaction score that most ERs use to assess how satisfied patients are with their visit. And we looked at it, and they were some significant findings. 

So it turns out that patients that saw the board, they felt like they were kept more informed about delays, and also, they felt like they knew what to do after discharge. And this is a really important thing because as a ER doctor, I think about often, do my patients understand what I told them to do after discharge? Do they know, refill this prescription? Do they know when to come back? And I think that's one of the biggest things in emergency medicine because sometimes there's a lot of uncertainty in emergency medicine for the patients and the doctor. 

And you don't want to send anybody home that's really sick, but sometimes you have to have that discussion and just be like, look. If you feel worse, come back to the emergency department. And I think having a tool that helps them understand their visit, and they, afterwards, can say, oh, yeah, I feel like I know a little bit better what to do after a discharge, is really good. And having a tool that explains delays or lets them know where they are in the process of just their emergency medicine visit can help them be more engaged, get better outcomes, ask better questions. 

BRENDAN: So what was the inspiration behind that project? 

ANDREW MARSHALL: So I think we were-- basically, the Brigham was approached by E Ink, which is a large corporation. They make e-paper screens. Those are used in Kindle screens. And they said, we have this great product, and we want to use it in health care. That resulted in us getting linked up with the company as providers and hopefully innovators. 

And so we thought, well, wouldn't it be great if patients could see where they are in their visit? Because I spent a lot of time in fellowship thinking about this problem as well. How do you reduce anxiety? How do you engage patients? And then once they're engaged, how do you display contextual health information to them? 

Because I think what we're seeing now is that a lot of-- now that people have instant access to their records, there's not always context as to what the results mean. Is this radiology report-- this incidental finding. Is this something scary I should be worried about? Is this lab value that's 1 point over something I should be worried about? 

And so it's creating a lot of questions. So I think being able to, during the visit, provide the information to them with context was something that was really exciting to us and so I think a really cool way to use this technology within medicine. 

BRENDAN: Yeah, that's great. I think that example of the E Ink displays is a great example of seemingly small things that can have a big impact on people. 

ANDREW MARSHALL: Yeah, for sure. Recently, there was a 21st Century Cures Act passed, which says that everybody-- there's no information-blocking. Everybody has access to their health care record. But if you don't have a device and you're in the emergency department, then we've got to figure out a way to get you that information because the people in the room next to you that do have their devices are checking it. 

They have better access. They have an advantage. And so to bridge those gaps, we've got to think about, how do we equitably distribute technology, and how do we do it in a way that removes the barriers of health literacy? 

BRENDAN: Great. Well, Dr. Andrew Marshall, Instructor of Emergency Medicine at Brigham and Women's Hospital, thank you very much for joining us on ThinkResearch. It was a pleasure to have this conversation with you. 

ANDREW MARSHALL: Yeah. It was really nice chatting with you guys. Thanks so much for having me. 

BRENDAN: Thank you for listening. If you've enjoyed this podcast, please rate us on iTunes, and help us spread the word about the amazing research taking place across the Harvard community. 

OBY: To learn more about the guests on this episode, visit our website, catalyst.harvard.edu/thinkresearch.