In this episode, we’re pleased to welcome Dr. Steven Rube, Chief Clinical Officer at IMO, to discuss combating healthcare industry headwinds through data standardization.
Podcast (hfppodcast): Play in new window | Download
Learn how to listen to The Hospital Finance Podcast® on your mobile device.Highlights of this episode include:
- What is IMO
- Health IT
- Data quality issues
- How to set providers up for success with complete and consistent data
- Physician burnout
Kelly Wisness: Hi, this is Kelly Wisness. Welcome back to the award-winning Hospital Finance Podcast. We’re pleased to welcome Dr. Steven Rube. He joined IMO in 2013 and now serves as the Chief Clinical Officer. He contributes a frontline user’s perspective to IMO’s executive team. He also leads a team of clinicians and non-clinicians designed to take a proactive approach to customer service and sales, both in the US and internationally. Dr. Rube has served as faculty at both Northwestern University and the University of Illinois medical schools. Prior to joining IMO, he practiced family medicine for 15 years in Chicago’s Lincoln Park neighborhood. He also served as the Chief Medical Information Officer at a large urban hospital in Chicago. He attended Case Western Reserve University, The Ohio State University College of Medicine, and Northwestern University’s family medicine residency program. He is board certified in Clinical Informatics and is a Fellow of The American Medical Informatics Association. In this episode, we’re discussing combating healthcare industry headwinds through data standardization. Thank you for joining us today, Dr. Rube.
Dr. Steven Rube: Thank you so much for having me.
Kelly: Well, wonderful. Let’s jump in, shall we? So can you tell us a little bit about IMO and what you do there?
Dr. Rube: Absolutely. IMO has been around for about two and a half decades, actually. Approaching our 30th anniversary shortly. And it’s the company that is in play everywhere, but very few people know about. IMOs original mandate when they started out was to be in the EHR space, the electronic medical record space, and enter that very competitive field. What was discovered very, very early on was that while the need to adhere to certain standards, coding standards for billing and for charging, there really wasn’t a standard language for clinicians. Doctors, nurses, therapists, and so on. The people at the front line. There was no standard language that spoke in the way they needed to at the level of detail that was needed by the clinician in front of the patient to capture the situation and to capture their clinical intent. So, what happened was you ended up with two very dissatisfied arenas. You had the administrative world who required code sets, whether it be ICD or SNOMED, in order to make these systems function properly because that’s how they were designed. And they were in clash with the clinical world that was trying to say something accurately about a patient that these code sets were really never designed to do.
So, IMO created what they called interface terminology. And the interface is really those two worlds between the clinical world that needs to say something often in a very finely detailed way and the administrative world that really just needs to capture codes and put people into categories for various reasons, whether it be reporting or analytics or billing. IMO came along and was able to satisfy that need on both sides. How we do that is a bit complex, but simply we have a group of clinicians, and we have a group of coders. So, we kind of pretend, if you will, that we’re the real world, and we map those terms for them. So, when we deliver them to all of the major EHRs in the United States and even in the English-speaking world outside of the United States, both sides are able to accomplish what they needed to do very seamlessly. The clinical people say things that they want to say in the language that they were trained. And as that data gets moved down the line, it seamlessly travels with all of those necessary codes that are required to keep the lights on. So hopefully that was fairly straightforward as to what IMO does. As we played in this field, though, we realized that there were other areas in the clinical data space that we could affect, change, and create products for.
Kelly: Sounds very interesting. So, what made you decide to get into health IT?
Dr. Rube: Do we ever decide is the question? When I came out of my residency in family medicine, I decided with a friend of mine who was also in that program to open our own practice in the city of Chicago. We decided that since we were starting from scratch, we were going to be the most technologically advanced practice in the city. This was when informatics was still in its infancy and EHRs it was like the wild west where there weren’t a lot of rules. We set out to do research to find out about systems. And what happens then is people around the hospital and hospital administrations, they start to take notice that maybe these two individuals understand a little bit about electronic medical records. We did do that. We set up our office that way. In certain ways, it was a success. In certain ways, it was very much a learning exercise. We were so bold as to actually build an office without a records room. We were under the misconception that we’d be able to do completely without paper charts. And this was around 1994. So, a little bit early. Not to date myself.
But therefore, what happened is it’s kind of–I’m not sure if it was me choosing informatics or informatics choosing me. But along the years in addition to practicing medicine, I was always paying attention to what we could do, how we can use these new technologies as more of a tool to practice medicine than a hindrance, and allow them to assist instead of getting in the way. As that happens, you acquire a body of knowledge. People start to ask your opinions. You run into situations around– for example, during Meaningful Use 1, the hospital that I was currently working at, wanted to use some of my expertise to help them get adoption of their electronic medical record. So, it was kind of an organic process. About a decade ago, when I ran into IMO, I thought it was very interesting, and started working here while I was still doing some clinical work. And I started, as I say, in the mailroom, creating content and really understanding what structured clinical data looks like, and we could get into what that means, as opposed to unstructured clinical data. Started getting into that. And then relatively recently, our academic arm, AMIA, established a fellowship for people who are interested in getting board certified in informatics. And I took that course and that exam, and then here we are today. Where I am now, as you said, working with IMO to really help assist clinicians in the struggle to capture quality clinical data.
Kelly: Very cool. What a great story that is.
Dr. Rube: Thank you.
Kelly: Yeah. So, IMO recently published findings from an industry survey that found that data quality issues continue to be a key issue for providers, with a surprising 98% of providers openly acknowledging that their organization must improve the way it leverages data to confront challenges. So why is this still such a big issue? And what are we doing wrong?
Dr. Rube: Interesting. I’m not sure I would label it as necessarily doing something wrong, although we could debate that. It’s a question of, the field was in its infancy when it was introduced into the clinical world. And the electronic health record companies that are out there are all doing a very good job, but what they viewed as the entire problem was really only probably a slice of the problem. And I’ll give you an example. So, the electronic health records are very good at capturing structured data. If I’m having a conversation with you, and you say, “I have disease X,” and I record you have disease X. We could capture that in a structured field. And great, everybody’s happy. I think what we learned over time is that so much clinical information is in disparate sources. It’s in other places other than captured in an encounter between you and a clinical representative, either a triage nurse or a physician. So, for example, if you have a procedure and a surgeon or a medical sub-specialist dictates a report, that information will be in the record, but it might be in what we call an unstructured form. Many people have health information in nonmedical applications, such as Facebook or LinkedIn, or other profiles that are contained on smartphones or smart apps.
So, there was so much clinical information out there, many of it, contradictory. A lot of it not being curated in any way. That was being thrown into this data swamp, if you will. People use the term data lake and data warehouse. I think they’re being generous. I think when you start looking especially at some of the larger institutions, or what we call IDNs, which are systems of more than one hospital, they may have different electronic records that don’t even speak to each other well. So, you really end up with this murky picture of clinical data everywhere, from dictated reports that the computer can’t really read or decipher to just strings of numbers and letters that are supposed to represent diseases. And that continuum really creates for an environment where data is not only not clean, but it’s not standardized or normalized. And that is one of the places where IMO really thinks we could help because, at an atomic level, we understand what these clinical terms mean. And it’s just a question of mining them, adding structure to them, and then you can go from data swamp to data lake, to data warehouse, where everything is now nicely folded and put into neat little drawers, and then you can use them as you see fit. So it’s not a question of what we’re doing wrong. I think we’re just still evolving in this process. We cleaned up one aspect of it, and now we have to start looking at all the pockets of unstructured data that are in what we call the data quality chain.
Kelly: It’s a very interesting way of looking at it, yes. It’s no surprise that clinician burnout and a looming recession are at the top of healthcare providers’ minds this year, as is everyone’s. So how can complete and consistent data address these issues and set providers up for success?
Dr. Rube: Sure. That’s kind of the Holy Grail that many are searching for right now.
Kelly: Right? [laughter]
Dr. Rube: Right. So yes, physician burnout is a problem. That’s probably a topic for a whole other webinar because it’s multifactorial. But we could address the issues that are germane to our discussion in the sense that regardless of what else is going on in medicine, now we are charging or kind of demanding if you will, that our clinicians become data entry people. And they’re not trained to do that, not good at it, nor do they see it as a necessary addition to the care that they’re trying to deliver. I truly believe that most clinicians walk into a patient room with the idea that they’re there to help alleviate pain or treat a medical problem. And the idea that they’re going to be tied to this software application and spend most of the visit with their back to the patient trying to type – I’m a two-finger typist like anybody else – becomes very, very frustrating. So, on a bigger scale, a way to look at it is we need to take these new technologies that– almost no one is arguing that the electronic medical record is not of benefit. It is. Coming from someone who was alive during the paper chart days, and I can’t tell you how many hours I spent looking for charts that were in radiology when I wanted to see a patient or trying to decipher handwriting, including my own, which I can’t even read half the time.
So medical records have brought us very far as far as patient safety and delivery knowledge. But what we really need to do is continue to strive to make it easier and more seamless – is a keyword – for the clinicians to get information into the system, and then on the flip side of that, the same. Get it seamless and easy for the system to provide information to the clinician. So, what do I mean by that? If you and I were having a conversation, the best thing the medical record could do is to be an application in the background. That it’s not intrusive into our discussion, into the trust that we have with each other. You’re seeing that now, and we’ll get into AI and NLP and machine learning, and people can argue about what all those terms mean, but can we use those tools to really assist clinicians into capturing accurately what’s happening at the point of care? Because we know that is the most accurate time and place to capture what’s going on. Not later, not having someone going back and sift through stuff that we’ve captured. It’s at the point of care we need to capture that accurately and preserve it. We then need the system to learn from that information that it’s capturing and then be able to assist in the workflow in parts of the workflow where it makes sense to assist clinical people in decision support and analytics and other types of reporting and identifying patients for clinical trials. All of the stuff that really slows us down right now. Medicine is getting more and more complicated. I was recently at an AMIA conference here in Chicago, and we were joking on how medical school is going to have to go from four years to six years to eight years if we’re going to expect new grads to absorb all of the information we’re asking them to. In order for that not to happen, we really have to provide them tools. Tools no different than a stethoscope or a blood pressure machine were 60 or 70 years ago. They were just tools. They didn’t take over what was going on. They just were another tool in the quiver that clinicians could use to practice medicine more effectively.
Kelly: That makes a lot of sense. For the last–
Dr. Rube: The shorter answer is we need to get out of their way.
Kelly: Yes. Yeah. Yeah, exactly. So, for the past few years, you mentioned AI already. AI has been a buzzword in the healthcare industry due to its potential to drive meaningful change in healthcare. Surprisingly, the survey found that 85% of provider leaders think AI has received too much hype. So, what is it that AI is missing?
Dr. Rube: Probably, again, another topic for an entire podcast.
Kelly: Right? Yeah. Exactly.
Dr. Rube: Right. So, what it’s missing is clarity. And what I mean by that is, again, if you ask a room full of 100 even informaticist– and this happened at this AMIA conference. In a room of 20, we couldn’t agree what AI was. So, I think there’s three main reasons why people are still skeptical about AI. One is that it’s not well-defined. So, if it’s not well-defined, how can I give you my opinion on it? Is an EKG machine that gives me an interpretive reading AI? I would say maybe in its primitive sense, but that’s not what we’re talking about today. So, the fact that it’s having trouble people defining it is a problem. Number two, and that kind of stems from number one, is that clinicians have a tough time visualizing what AI is going to do for them during their day. I practice medicine on a daily basis. I have a routine. How is this going to fit into my routine? Is it going to be another thing that’s going to slow me down that I’m going to have to learn? There’s that adoption curve that we talk about. Yes, there will always be a few people out there ready to adopt it right away. But for the most part, clinicians tend to be a cautious bunch. And so adopting new technologies that aren’t well-proven and well-defined is going to take some time.
I think a third answer is doctors are already again skeptical of things coming between them and the patient. Is this going to be another barrier between them and the patient that’s going to, again, erode that trust that patients feel when they come in and speak to us? And the last one I think is kind of the whole Skynet Terminator kind of thing or the machines going to take over. For us practicing medicine, there’s a bit of skepticism there, obviously. Not to that extreme, but physicians really still view medicine as an art and not an algorithm, if you will, that can be just marched down and spit out with an answer. So, we really have to be wary of that. I think there’s no question that these things are coming.
If you would have gone to an AMIA conference, probably two years ago, there might have been one or two courses or lectures on AI and machine learning. I charge you to find one course or lecture this last one last week where the topic didn’t come up. It came up everywhere. The exponential growth of programs like ChatGPT and just NLP ideas, in general, are progressing at such a pace that it really is upon us as an industry to get out in front of it, define it. Define everything from how it will be deployed to the morality of it, for example. I mean, are you obligated to inform patients when a doctor’s being assisted by AI? It’s very similar to when the da Vinci robots were introduced in surgery. It took a long time for people to get comfortable with that, as well as get trained with it, as well as patients to get comfortable with that a robot was going to do part of their surgery. So, I think this is like that. I think we can do it. But I think we need to get out in front of it because it’s moving a lot faster than a lot of people are aware.
Kelly: Yeah. No, those are some really good points. Very interesting indeed. It’s no secret that we have a data issue in America. And there are many companies trying to fix that issue to improve care and outcomes for patients. So, what makes IMO unique?
Dr. Rube: IMOs unique, I think– and this kind of branches off one of the questions I answered earlier–is that we started in a different place than most other companies. We started at the clinical language level. Most other companies look at the standard code sets, ICD and SNOMED. And this is no criticism of those. They do very well what they were designed to do. ICD, the International Classification of Disease, it originally was International Classification of Death. And that’s not a joke. It was a code system created to put people in buckets. You had an illness for reason one, two, or three, a different reason, or I don’t know. It really wasn’t about specifically calling out what was going on with the patient or documenting it. SNOMED is probably the international leader for a clinical code set that’s designed for analysis, collation of data. But it’s meant to be used after the fact. Once you have data collected, how can you categorize it in a way using an ontology to really analyze that data? It’s written in their charter that way. No one went to the clinician and said, “We’re going to create terms to say what you want to say, and then we’ll apply the data on the end.”
Now, why do I give you that history again? Because that means when you’re trying to clean up the clinical data chain and you’re trying to clean up these swamps and turn them into cleaner versions of data and normalize this data. IMO was really the only one who was there for the birth of these terms. In other words, we understand what clinicians are trying to say at a subatomic level. So, if you broke everything down in medicine, we were probably there. We probably helped create the terms that those clinicians used at the beginning, and then got dumped into a data swamp and a lot of the information was lost. If you think about the fact that IMO is right now in about 85% of the acute care facilities and all of the major EHRs, people are already using IMO’s data to capture what was going on at the point of care. We just need to follow that through the data quality chain and make sure that information is not lost. Then, like a magic decoder ring, we can look at the big data aggregators and say, “You know what? We recognize a lot of these terms. We know what that means. We can read an unstructured piece of text and understand what the clinician was trying to say because we’ve been speaking their language for almost 30 years.” So I think that’s the major aspect that really makes us different than some of the other companies that are looking at data aggregation and normalization.
Kelly: Sounds good. And so how can someone learn more about this podcast topic and/or IMO?
Dr. Rube: So probably the best way, one of the best sources of resources for IMO is our website at imohealth.com. There’s not only information there about the company but there’s a lot of articles by people similar to myself. I happen to have on my team several nurses and physician informaticists who do podcasts, who write letters. We have most of those resources available on the website. And it really is interesting with all the different backgrounds that we’ve accumulated over time to have the different perspectives on clinical information that we have within IMO. We’re also very available for people to reach out to. My email: srube@imohealth.com. You could reach out to me. We have many people. We do a lot of webinars and seminars. And we’re very passionate about clinical data and the use by clinicians in the real world.
Kelly: It sounds like you are, for sure. So, thank you for giving us that information. And really thank you so much for joining us today, Dr. Rube, and for sharing all this fantastic information with us.
Dr. Rube: No problem. It was my pleasure. Thank you very much.
Kelly: And thank you all for joining us for this episode of the Hospital Finance Podcast. Until next time…
[music] This concludes today’s episode of the Hospital Finance Podcast. For show notes and additional resources to help you protect and enhance revenue at your hospital, visit besler.com/podcasts. The Hospital Finance Podcast is a production of BESLER | SMART ABOUT REVENUE, TENACIOUS ABOUT RESULTS.
If you have a topic that you’d like us to discuss on the Hospital Finance podcast or if you’d like to be a guest, drop us a line at update@besler.com.
