As one of the world’s fastest-growing interdisciplinary fields, healthcare informatics continues to revolutionise the healthcare industry. There is a demand for knowledgeable and experienced staff and faculty in healthcare informatics that far outpaces the number of available trainees.
Dr. Samuel Volchenboum, MD, PhD, Course Convener, and Suzanne Cox, PhD MP, Head Tutor on the University of Chicago Healthcare Informatics program, discuss the different career paths and professional interests of students on the program and answer some questions from the student community.
Watch the Video with Dr. Samuel Volchenboum, MD, PhD and Suzanne Cox, PhD MP Below
Click here to read the full transcript
Suzanne: Welcome, everyone. I think most of you will probably be watching this as a recording. Sorry for the late notice. We couldn’t get the link out until yesterday. But we hope that you find this a nice addition to the class.
So I just want to introduce Dr. Sam Volchenboum, who some of you have seen in many videos, and who has been along with us on our learning journey for the last eight weeks.
So Sam, the classes have been going really well, and the level of engagement this time around has been really high. This course has been amazing, people have put a ton of effort into their discussion sections and doing their assignments, and really engaging with the class. And so it’s just been a terrific presentation this time.
They’re very passionate about a lot of the topics that we covered, and I think I shared with you a lot of the students have already been talking about how they’re bringing it back to the workplace now and how they see their crew moving on the future. So it’s just been a really terrific class.
So I wanted to start off in community sessions, by just talking a little bit about the original theme that we start the class with, which is identifying for the students different career paths and professional interests that they might have, and how they can think about that by using this biomedical informatics matrix that we provide for them at the beginning of the class, and then towards the end of the class we start leading back to that. They learned all of these different topic areas as we’ve gone along.
I wanted to give the students the opportunity to hear a little bit more about you, Sam, and how you first became interested in informatics and how your interest in informatics integrated with your medical training.
Sam: Yeah, great question. Everyone takes their own path, their career path and I was no exception. And remember, I grew up in a time when the idea of becoming a programmer or using computers for medicine was not on anybody’s, or not many people’s radar. So as as I went to college I had an interest in using computers, but at the time most of the applications we’re using with computers and medicine, were relegated to things like finance and administration.
So for me, as somebody who wants to get into medicine primarily, that my path was very directed into going to medical school, and becoming a physician. Not becoming an oncologist, and you know in a way things are much better now because if you want to leverage informatics and medicine at the same time, there’s much more defined career paths for you.
So for me, I went through my medical training, but I always found that I was very frustrated with the inefficiencies of care that existed. I was just talking to somebody in class about this the other day where the first day I showed up for a fellowship and I wanted to order an x-ray on a patient.They told me to fill out this form and then fax it down to radiology, and about the third time I filled out the form and faxed it, I said, this is ridiculous. Why are we all faxing these forms? I walked to radiology and I said, “Hey, if I emailed you this information, would you still use it and they said of course we would take the information and would prefer that.”
So in, as you guys are already thinking, I just went I created a web form people all over the hospital started use in this form to fill out very basic aspects of radiology requests, and they would hit a button and it would go to the inbox of radiologists. Now, this is not an Enterprise System. It wasn’t sustainable.I made a lot of people angry who were working on Hospital systems at the time. But it allowed me to start thinking about how to improve efficiencies of care through informatics.
So at that point, in addition to starting to build up more of these tools that helped people. I went ahead and decided to formalize that education through my master’s program in biomedical informatics.And the the initiation into that program, the topics are very much the same things that you are learning in this course. It’s understanding about Hospital Systems, about databases about about using informatics tools to help improve Healthcare and that actually having that training actually then allowed me to use both medicine and informatics to help build a career right now actually, you know run a group that uses informatics tools to help improve care.
So for me, it was very much, two parallel processes that converged but I think for those going into the field now, it’s much more of a organized process where you can actually understand the kind of field you want to get into and take the right steps to deliberately govern there.
Suzanne: So, during that first week of class, because they showed themselves for business students that they are. Some of them are already finding articles on their own and posting them on the discussion boards, and there’s a lot of conversations that went back and forth and somebody posted the white paper that defines healthcare informatics and its core competencies talked about research and education and it was interesting for me to review that paper again.I looked at the figure that, let’s take it from the textbook that we use and it summarizes the discipline and the areas of practice in a similar way that we do in this Matrix that we use in this course.
They added this cross-section that cuts across all of the areas in translational research, and I thought that was interesting way to present it.So can you like talk a little bit to the students about translational research and how it cuts across both bio-informatics and clinical research? And maybe a little bit about what you think the future holds for translational research or why it’s important.
Sam: Yeah. So obviously we talked a lot about translational research and it drives a lot of what we do and loosely defined you think of translational research as, you know, the old cliché as bed to bedside, right? You want to take something that’s basic research and you want to figure out a way to take that very easily into the hospital and to take care of patients.
I think that’s a little good old-fashioned definition now because now we really want to think about how do we use basic science discoveries as part of a learning Healthcare System where we continually drive forward the discovery into patients in the use of knowledge we gain from patients back into discovery again. And I think what’s interesting for me and looking at that Matrix is that the old traditional paradigm of thinking well if I’m going to be a bioinformatics professional, this time doing basic science research, that’s quickly, we’ve quickly broken that down now where you could be in any area of informatics research, whether it’s hardcore genomics research up to population sciences and still be doing very translational work.
And I think a lot about, my genomics program here, where we run genomics pipelines to understand about germline mutations in patients, and we feed those results were right back into clinical care, almost immediately. And there’s an area there where you know five years ago. I couldn’t even imagine being able to collect blood from a patient run genomic test that makes a clinical decision about that patient, and then learn from that decision all within a couple days. So I think the drive towards more translational research and the incentives for doing better translational research has really driven the science forward. I think it allows you to train these new trainees, like you guys in this course to understand that if you start to collect these skills and learn this area of science and being able to make part of this larger translational medicine.
Suzanne: So do you think that informatics is keeping up with the speed of Discovery?
Sam: Yeah. I mean, I think it’s interesting, because in some ways informatics is far surpassing what we can do from a translational research standpoint. We have much more data than we can ever hope to analyze right now. We have a lot of computer power. We have a lot of ways of analyzing data, but we don’t even have the systems in place to take advantage of those things. So we we might be able to use deep learning to predict who’s going to get sick in the hospital. That’s actually an interesting scientific informatics question.
But how do you logically create a system where you can implement an algorithm safely in a hospital? That’s a governance issue, but we have a long way to go to catch up with those governance issues to the actual science we’ve done. So, when you say the pace of informatics is both the scientific pace, which I think is moving along and breakneck pace, and then there’s the actual implementation science which is where I think some of the new breakers are really going to have to happen if we’re going to push these things into the clinic.
Suzanne: So also among the academic medical centers in the US and I’m working in clinical and translational research, there’s this big push to standardize research data. So that investigators need to collaborate better. So will you tell us a little bit about data standards and some of your work some of the consortio that are creating data standards and what that’s been like, right?
Sam: So, you know, a lot of Industries have really been able to propel themselves for by deciding on data standards. It would it would seem crazy if you went to an ATM now and your ATM said I just can’t operate with most of the banks out there because I’m using a different standard.That’s not how it was, you know, 20 years ago. And so I remember when ATMs first became popular that you had to go to one that was just for your bank.
So, much in the same way, medicine I think is 20 years behind the bar right now because most of the data that we collect is done so in a one-off way that’s very customized to that particular hospital or that particular environment, you know one area that I’m very interested as clinical trials data. Clinical trials data are collected often still on paper forms or in ways that are very customized to a particular investigator or a study.
That’s when it comes time to try to share those data, you have to then convert it to standard so that everybody can share the data and that conversion process is very time-consuming and sometimes impossible depending on how the data were collected. The same is true for clinical data. I was talking to somebody the other day that we don’t, you know where I’m at, we don’t collect our pharmacy data in a way that’s standardized. We have our own internal standard. But the National Standard of using rxnorm is not something that we’ve been able to convert to. So if we want to share a pharmacy needed with another institution, somebody has to do this very painful converted process.
So we’re seeing this across the industry, and the problem is that there has not been a lot of traditional incentives or using standardized data models because we did not have a lot of incentives for sharing data. That’s really changing and I’ve seen a tremendous shift over the last couple years. So that when new systems are being built and implemented, there is a real push to use this to standardize data collection methods so that we can end up sharing the data.
So an example from my own work is looking at clinical trial data for pediatric cancer. Where these data are collected traditionally have been collected in a very haphazard non-standardized way. So my group here now has been going around to the different pediatric cancer groups for trying to get them to understand and converge on a data standard so that we can actually collect the data and share it. So I think there’s a tremendous opportunity here. I would love for people went into Data standardization and more people wanted to understand about ontology dictionaries.
That’s a very passionate area for me. And I think that’s actually going to be one of the things that drive Health Care forward is is a better a better approach to data standards and harmonizing data.
Suzanne: So, it’s interesting, we share this film to our students here, No Matter Where, which is documentary about trying to get hospitals across the US to share your data and have. So that one hospital can pull up the record of another patient. There’s this great line in that film that says, you know, there’s no reason people don’t understand why when they show up at 7-Eleven. Person behind the counter knows more about you than somebody at a hospital. For those of you outside, the 7-Eleven is like a convenience store or something where you just go buy gallon of milk the best sort of thing.
So, do you think that there’s a role? For consumer advocacy of consumers and patients are really saying this unacceptable. Hospitals need to have my information when I show up. Will that be part of the driving factor for moving data centers in the clinical setting forward or where you think the real question is can have to come policymakers or conditions or combination?
Sam: Yeah, you reminded me of a funny video the YouTube video where guys trying to, it’s a farce about trying to air travel, and, it’s somebody on the phone, saying you have to fill out that form He’s saying I filled out that form 10 times. It’s the same form. Why can’t you look it up?
And the whole point of the video is to is to point out how ridiculous it is that we can’t share even the simplest kinds of data, and I’m sure many of you listening has had the experience you go into positions office you fill out on the farm often on paper, where you’ve answered the questions, maybe even maybe even that day.
I think there is, a greater frustration from consumers now to try to demanding more standardization of data, I think we’re seeing, I think we’re seeing more consumer advocates consumers advocate for their own, trying to optimize their own interactions and the Health Care system. And I think as more and more people take control of their own health care data, as people start to own their own data. As the EHR starts to move out of The Silo hospitals and more into controlled environments or people have access to the raw data, I think we’re going to see much more of a standardization of probably collected the data in a much better sharing of data so that you could walk into any hospital they could identify you and actually have your records available for you.
Now, you know that seems like technologically we should be able to do that now, and we probably do it from a text standpoint, but it’s going to take a while before we get there the US. You know, even in the US we’ve had a very difficult time having the discussion around Universal identifiers. People start to get nervous when you start to attack them with the identifiers so that one number can look up everything about you. And in the European countries there even more restrictions, so it’s going to be allowed when we get there, but we have to obviously because the current system is not is nothing run up against a wall and how much we can get out of the.
Suzanne: Yeah, I think it’s interesting. I think a lot of it is because in the US and other countries, you know the systems were set up sort of an isolation. But some of our students are moving in places where systems today are just thinking about how do we roll out electronic records in our country and I think you know much the same way that we’ve seen countries that sort of Leap Frog over biotechnology and just having Wireless technology where mobile phones with them sort of first universal connected connected devices, maybe some of these countries have opportunity to actually surpass that, and have connected systems from the beginning.
I think some of our students have real needs to be part of that actual. So, okay, so following on from that, Lee asked a question about software. So [00:15:00] yes an effort to reduce the cost of healthcare some argue that software should be open source as opposed to proprietary. And he was wondering about is it did that to Innovation? So maybe you could talk a little bit about software sharing and research. But then also extend to other incredible applications.
Sam: Yeah, you know, I think of course in the research world, there’s a there’s a strong move toward open source software and that that is not only because most funders require that you share software they can develop in the course of research, but it’s also to provide a lineage and uh data provenance that you actually have a record of the song what you developed, and how you did your analysis.
So whenever we do a research project, it’s always with an eye toward knowing that we’re going to open source software available. I think in the clinical world is much more difficult. Because you’re building systems that have that have a liability that have to conform to a very high standard for consistency in ability.
And so much of what you’re getting when you pay a billion dollars or epic installation or server installation or any commercial EHR you’re paying for that reliability and you’re paying for that support that comes along with it. In an open-source community, of course, you can get you know, you can get very high high quality software that has a huge group of users, but it’s often hard to get that same level of support. You could imagine a model where there’s an open source platform and then that gets commercialized for the installation. And we’ve seen that in other areas where um an open source platform then is available and then a company takes that platform and developed a commercial product around in verbal imitation.
But I think with with commercial EHR software it’s hard for me to imagine that the software itself will become open source. What I do see happening is a lot of open sourcing of materials that can be used as part of the clinical care. So for instance, I know epic the best success of be easier.So epic has smart forms, which are data collection tools for medical care. And when you develop a smartphone you can actually make that smart form part of a larger community that other groups can use that form, so that you can standardize the data you can bet the form and then other groups can use that form for other standardized data collection.
And I think we’re going to see that more of that concept when it comes to tools and data collection mechanisms and forms and uh server has these these, these App Store concept. I think that’s what that’s what we’re going to see much more of the open sourcing but that doesn’t mean we couldn’t see the emergence of a larger resource EHR platform. It’s just hard to imagine in the current environment that there would be something that would take over, where there’s already, you know 100% Market penetration for commercial vendors.
Suzanne: Okay. So what we covered the US Healthcare System students share their experiences with their own systems their country is mentioned and they talked about what words what’s not working and is a discussion groups. They talk a lot about shift to die you based care in the United States.So what’s your take on value based care in effect on having as having both on care providers and also uninsured?
Sam: Yeah, and you know, I’ll be the first to admit that I’m not, I’m not in the clinic all the time. So I’m not necessarily impacted by um by these changes my day-to-day work. What I will say is that that additionally we taking care of patients in a way that allows us to order whatever tests and procedures we need to take care of the patient.I think in some cases that could be abused, and that can raise the cost of Healthcare and so it’s in some cases I think shifting to a value-based model model or a bundle model makes sense and concertedly drive down the cost of care and try to decrease the cost of unnecessary tests, that perhaps were being used to increase reimbursement.I think there is a danger in doing this in that we are going to be creating a system where all we care about documentation.
And we’re still fixated on checking off the boxes. Before that, were afraid to order a test that’s going to increase the question mark, even though we think it’s medically indicated.I think there has to be a balance here, and I think if the shift to thinking about value-based care has been helpful instead is getting people. You think about work for the knee replacement surgery the after here preventing infection and the rehab as a single as a single entity rather than traditional model where you would have, you know an orthopedic check-up, and you have a surgeon that you have somewhere else where those are all just different parts of the care.
So I think it’s really allowed us to make more about the full the whole patient’s journey, which I think is very helpful. I think it’s, I think it’s not helpful to think about everything becoming part of some bundle where we can understand exactly what it’s going to cost to take care of condition. I think the other important shift is really understanding. The other part of value based care is the value Mark which is understanding. How do we measure the changes that we’re doing? How do we measure the impact on health care? So there’s there’s been a real shift even in the last couple of years I’ve seen a real shift to try and understand how to remember the quality of the care that we’re doing.
It’s not just about that individual patient outcome and understanding what our infection rates were our efficient rates and by tracking those things better. Even if we’re doing it through this rubric of value-based care funnel care model by tracking those things better we’re going to understand about the interventions we make and how to how to get better care at a lower cost so I think overall it’s a good it’s a good shift. We just have to be careful what the approach is.
Suzanne: Yeah. I think you know it really delete question and all seem to talk about it. You know, I was thinking too about my perspective which is uh that’s running the master’s program here in biomedical informatics. I have a lot of different companies come to me and ask for students to volunteer on projects. And we’ve seen requests across the board. So we had insurance companies or people who work with insurance companies come to us because they’re trying to use data analysis to determine whether or not the cost that they predicted for Edition bundle is going to work out or whether or not this diagnosis group as they defined it fits into their business model.So that’s one way that informatics is really playing a role in evaluating.
But then also from the providers we have hospitals coming to us concerned about maximizing the efficiency and care as you talked about, but then they also have to be really careful about their ecology metrics. So they’re using a lot of predictive modeling, um, try to identify patients who are risk Adverse Events and proactively manage their care so that they can make sure to try to minimize those negative outcomes. So it’s in that way, I think it’s really working that’s part of the ideas, too.
Raise the quality of healthcare.So they’re looking at you. We think might you know in our current system would be at risk for readmission or my need additional surgeries because of something you have been at higher risk. So both of those who count against you in quality report. So it’s who’s the hospital to get out ahead of that to those people might be interesting that those groups are coming because they need the data to help talk to their algorithms.
Sam: I mean that actually shows another problem in the processes that improves need to provide better care of. The only would have access to primary data to do that. Maybe there’s an opportunity there to help create a better environment where there’s a better feedback loop. So people can learn more quickly about the care.
Suzanne: Okay. So, Richard also submitted a question. So, first of all the course, thanks for doing this for you and he wanted to know Sam what you thought about the sustainability of how, for Champion Healthcare informatics and the opportunities. Because you share any classes learn about implementation and use service years?
Sam: Yeah, I think, I’m not quite sure exactly what, what exactly the thrust of the question is, but my impression over the years is that we continue to build a lot of one-off systems, and a lot of a lot of one-off solution. I think about her clinical trials management system approach where when we take out we take a pharmaceutical trial in one implemented and collect data on nations for a study, many of those systems are just built as one data collection and they have basically no scalability and no sustainability. And that’s because there’s been a lack of incentive over the years for doing things in a sustainable way.
The same is true for many different areas of Health Care clinical care, when we take care of patients. We often create a way to collect data for that patient that is not scalable or extensible to other areas. I think this is something that people are paying much more attention to especially as. Try to do things in a more caustic way. I think long-term sustainability is only going to be achieved through better data standardization for enterprise-wide approaches to development of software in tools, and understanding more about how we can share use data that’s been better ways to share a few structured data that are going to a certain standard.
This is a really exciting area because this whole Paradigm of collecting data only when they’re in the hospital, is this is really our very artificial when you think about it in the way we’ve collect data from patients at home has been through a paper and Pen surveys and sometimes through, you know, tablet or phone surveys. But general we don’t do much work in getting what we called. You know real world evidence, which is you know, how how is the patient doing outside of the hospital outside of their traditional clinical care? And the use of wearable sensors to collect passive and active data from patients, I think is going to transform both clinical trials and clinical care, you know, imagine and how right now when somebody wants to track our patient is sleeping.
They’ll ask them every couple weeks when they come to Clinic how they slept over the last few weeks or the pull out a survey you can imagine a system where people just use a wearable that tracks their sleep every night and allows the physician to actually track in real time applications doing and actually contact the patient if there are problem.
They can look at the population of patients that they take care of four different trends. It’s really going to transform it revolutionize our ability to take care of patients and the one quote I look I like that I heard was, all data, from all patients, all the time. Because that really means we’re going to be able to collect information from our patients wherever they are, of course with the proper privacy and security in place, but it’s think patients really want this they want to be able to have their data be used for their aspects of their care no matter.
Suzanne: Okay. So you had an article in Wired Magazine last year that address the promise of challenges with using things like social media and other non house data sources. So tell us a little bit about what holds um, and whether or not you think this recent probe into the Facebook policies and affect your view about that Trend since the article.
Sam: Yeah. I think this is this is a tremendously this is a tremendously important area, right because we have this digital exhaust that we leave everywhere every time we were shopping every time we log on to a website every time we do a Facebook post or Twitter post. We’re leaving little bits of ourselves out there that can be ultimately used used in ways which can be nefarious and we certainly see this recently in people very worried about the collection of data by Facebook and others.
But I am very optimistic that these data can be used in a very helpful way so that if people are on Facebook talking about symptoms related to their disease or even related to diseases that they don’t even know they have, there’s a tremendous opportunity there to to use those data to help take care of people to help make predictions about their health care.Uh, and I think if done correctly, nd with patients’ consent to move their dying and I think this could be a tremendous advance in how we think about the use of data for healthcare and again into getting away from the traditional model of using the data that are collected as part of the routine.
How visit is starting to decorate it with other data, whether it’s from wearable sensors or in this case from our exposure to other systems, and I’m not just thinking about Facebook and Twitter thinking about your credit card history and the things you buy at CVS and your GPS data from your phone. Yes, it’s very scary to think about if these data were to fall into the wrong hands what they can do but think about if used in a in a proper way and use it good, think about how amazing is it being to connect all these areas and actually start to make some real predictions about care and our entire Health maintenance.I think it’s a tremendously exciting area.
Suzanne: Okay. So this week we’re finishing up the course with animation entrepreneurship section. So just as the final question you talk a little bit about the trends that you’re seeing here. Um, and I know I personally see a lot of clinicians who are really interested as into entrepreneurship and developing products.Yeah. I mean again, this is obviously an exciting area where um where Healthcare technology is obviously, enormously impactful area with a lot of potential for for Innovation and Entrepreneurship and I see it in several ways. Obviously we see a lot of companies that need access to our systems and and thinking Partnerships with academic medical centers.
So we see companies approaching us all the time saying we have this cool. Product to develop need access to your data to develop it and we Partnerships and even cold leverage IP together to build an electrical property that we can then spin out and used to start new companies. So that’s an exciting area in and of itself.The other area which I think is probably more towards your question is there’s a lot of people that are in healthcare that see these opportunities to take what they’re what they’re studying and how they’re taking care of patients and spitting that out into into into companies that can then be used to scale to other areas.And I think this is uh, traditionally people thought of this as drug development as device development.
And of course, those are very important areas that continue to be very impactful. But what we’re seeing now is more of a trend of understanding. How do we take the advances that we create as part of the clinical care system in turn those into opportunities that can be scaled to much larger systems?So I’m thinking about, you know, we were developing ways to implement clinical trials better and better ways to standardized process. You know are there areas of clinical trials that developed into personal interests? Of course there are. I think about the ways that we run our bioinformatics pipelines that are genomic data.There’s a need to standardize those to create companies where those um, those tools and techniques to be leveraged across the industry.
What is usually missing is the infrastructure to take those advances from them into a commercial ice interest. And so the real blue here is having the intellectual property lawyers.It’s having the commercialization folks that really understand how to do this available to researchers and and University of Chicago, I think we’re we have a great Paradigm. We have to business school. We have the technical technology transfer folks. We have the entrepreneurship protocol co-located where they can work with Physicians to help spend our company.So this is a very exciting area. I think it’s, I think anybody who’s a good a degree or certificate in biomedical informatics is obviously our rethinking about these opportunities and I would love to explore these areas with anybody who wants to have further conversations about. Okay great.
Suzanne: So, um, I think that’s the last anybody has an additional question that they want to type into the chat box. I think we’ve covered a lot of the different things that came up during the course of the last. Thanks everybody for listening. I think most of you are probably watching the recording after the fact but things for those of you joined by, I hope that you found this interesting exam. Thank you so much for joining us and we’re developing this course. I know everybody’s just gotten a lot out of it and it continues to really, you know, a lot of people and a lot of great International conversations and it’s been totally fun for Chris and I to the students again this term and, thanks.
Sam: Great. Thanks everybody for coming on and forward to their interactions. Okay. Thanks.