From Data Nerd to Surgical Innovator: Bruce Ramshaw's Story

We had to come up with a new measurement. We called it the emotional complexity. High-medium and low. As we collected that data and we did the non-linear analytical tools, we saw it was the highest modifiable factor correlated outcomes. And it was like, that blew me away. It kind of humbled me because there was a factor that was really impacting outcomes that didn't have anything to do with my surgical technique. You're listening to data nerds in the OR, a search and journey toward value-based care. The podcast focused on data and data science as the keys to a better healthcare system. Ideas brought to life by the vision and experience of host Dr. Bruce Ramsha. Each week, Dr. Ramsha sits down with different players in the healthcare system to discuss how data is used today and what it could do if it were used effectively. How it could improve surgical quality, education, outcomes, and drive better patient care. Let's dive into the latest episode. Here's your host, Dr. Bruce Ramsha. I liked it last time when you just started with kind of your story. Bruce about how you got into surgery. I think that's a great place to start and into medicine and staff. Yeah, when you talk about data and stats, the first memory I have of being a data nerd was being at home by myself over the summer. And I was very much in the sports, but had a lot of time on my hands. So I would play out baseball games on the carpet in a living room using a little paper football. I'm looking at it and I'd had the whole stadium laid out. And I would record all of the at-bats, whether it was a hit or a single double triple home run. My favorite team was Boston Red Sox. My favorite player was Carl Ustramski. So whenever he was up, I would always flick it real hard and usually hit a home run. So his batting average was like over 900. But I would record these stats over many games over the whole summer and add 'em up. So that's the first memory I have where I really got nerdy about numbers. I know during the seventh grade, we had to do a science fair project. And the project I picked was NFL football injuries. But instead of doing traditional prospective randomized control trial, I did data science. I collected all of the stats that I could think of that had to do with potentially resulting in an injury. So there was things like the type of field where there was indoor, outdoor stadium, the type of weather when the injury occurred. What was the position of the player that was injured? And back at that time, this is 1975. So there were only a few games a week on TV. There was a Monday night football game. There were free games on Sunday. And when the games were at the same time on two different channels, I had my younger brother record the data on the other game that I wasn't watching. And I think some of the motivation was, so I could tell my mom that I have to stay up for the entire Monday night football game past my bedtime. But so at the end of the season, we had all this data we collected on NFL football injuries. And I crunched the data and basically saw that there were correlations between extremes of the temperature and the type of turf, artificial turf. And very cold, very hot and artificial turf were highly correlated with more injuries. And that was my science fair project. A couple decades later, the NFL brought heating and cooling technologies to the sidelines and replaced the artificial turf. But I had generated that information in my seventh grade science fair project. That was very cool. So that's a couple examples of some of the data that I was as a kid. What was your favorite football team if you were a Red Soxan? So I grew up in Southeast Florida. So there was no baseball team in the majors. And there were, I think it was NBC weekly baseball game on Saturdays. And for some reason, I thought the green monster, the left field wall in Fenway Park was a really cool thing. And Yastremsky was the left fielder. So I adopted them as my favorite team. But because there was an NFL team in Miami, often that was my team. And Larry Zonka was my favorite player. That's interesting. So how did you go from a sports loving kid who was into statistic to a surgeon? How did you get there? That was a long process. I really didn't know what I wanted to do. I had a good mom. She really taught me to help people make my life and my profession about helping people. So as I entered college, I thought about either being a teacher or a doctor. And during college, I volunteered at the local hospital. This was at University of Florida, Gainesville. I volunteered at Chan's Medical Center and did things like taking newborn babies out to the car and helping people get back into the clinic. Yeah, it was just really cool to provide care or a little piece of providing care for people. And so that got me interested in medicine. Fortunately, got accepted into medical school at University of Florida state that there, I went in very idealistically. I thought, OK, I did all the work required to get in the medical school. Now it's time to learn how to be a good doctor. I wasn't sure what type of doctor I wanted to be. But the first month, I got really disillusioned. It was all about memorizing facts and scoring high on tests. And the first test was an anatomy test. And people were complaining about the questions. And it just-- it really turned me off. It was so bad. I quicked on a class. I got a job at a bar as a door man. And I learned more about being a good doctor working at that bar for four years than I would have gone to class. I made $5 in an R plus. All I could drink on or off the clock, which was my first six-figure salary. But I really had to talk to people when they were in various states of anger or drunkenness. And instead of using kind of violence, I would always just talk to them, say, hey, you can't be dancing on the dance floor knocking people over. You need to calm down. And as long as you cheer them like a normal person, I never had a time where anybody started to fight or anything like that. It wasn't a really rock-ass bar or anything. But I was very shy by nature growing up. And it was helpful for me to learn how to talk to people in those situations. During third year, a medical school-- you have to do your rotations. So I went back and started doing my rotations, which was a lot better than the first two years. We were just memorizing facts to take tests. And I really enjoyed those rotations. I did pretty well. But I thought I was not going to do surgery. Because back then, this is in the 1980s, surgery was a very malignant type environment. The most of the surgeries were older white males. And they were typically pretty nasty to work with. Or at least a lot of them. That's where things like throwing instruments and screaming and yelling was not uncommon. There's even cases where the surgeon would get in a fist fight with anesthesia, just. And thinking back then, like, now today, that should be totally unacceptable. But back then, it definitely happened. I witnessed some of it. So I didn't think I wanted to go into surgery. So I put my surgery rotation last. But when I did it, there were some very kind resents that allowed me to do things like suture. But probably most importantly, I saw people suffering. And then I saw an activity that relieved their suffering in a matter of hours. If somebody comes in with a ketopenocytus, and they're really writhing and pain, and then you do surgery, remove their penics, and they wake up and they feel better already. It was almost like, if you're-- I think a lot of people go in this surgery, I think of the most tender-hearted medical students. Because they see somebody suffering. They see somebody doing something. And they see that suffering is relieved through that activity. It attracted me to that. Even though there was a lot of negatives in surgery at that time, in terms of culture and the environment, I tried for surgery. And I went around the country interviewing it. I think 25 places. I only found four out of 25 that I felt comfortable would be a decent environment for training. And I was fortunate I was able to match at one. And it was a small community program, two resents per year in Atlanta, Georgia, called Georgia Baptist. And that was the transition and kind of my experience going through medical school-given, very disillusioned, but then during this surgery rotation, really seeing how much you can impact a person's life through that activity of performing an operation. I never really thought about surgery that way, that you're immediately solving somebody's problem. But are there downsides where surgery goes wrong or you started to see negative outcomes that kind of led you to look at how data science applies to surgery? Yeah, I think the biggest impact I had early in my career was at this small community hospital, Georgia Baptist, just through a number of coincidences. I started in 1989. And in 1990, we did more laparoscopic surgery than any hospital in the world. And it changed surgery quite a bit because instead of the primary surgeon through an open incision where not many other people can see what's going on. Now we're doing surgery off of a screen where everybody can see what's going on. So, you know, if your skills and your complications are all up there for everybody to see. But minimally invasive surgery wasn't adopted easily. I remember the reason, one of the reasons why we did more laparoscopic surgery in 1990 than any hospital in the world was that the American College of Surgeoned Meeting was in Atlanta and our chair allowed us to go. And I remember the very first time a video of a laparoscopicalisostectomy was shown by Eddie J. Rennick was at the US surgical booth at that meeting. And one of our fourth year residents, David Barrett, saw it. And he thought that there was going to be something to that. That was going to be important to learn. And I remember other surgeons going by, we're just making fun of it. I remember one called it Mickey Mouse Surgery or something like that. And so David Barrett asked our attendees, "Hey, can I go learn this?" And they said, "Yes, and we'll all work together," which was very unique. Most places, the first laparoscopic surgeon either by themselves or with a partner did it and everybody else was against it. But at our place, all the attendings were supportive. And he went up to, David went up to Eddie J. Rennick and said, "Hey, can I learn this somewhere?" And he said, "Yeah." Not many other people want to learn it yet. So he spent a month with Eddie J. Rennick in Nashville and then brought that back in in 1990. That's why we did so many laparoscopic procedures. And that became the model for learning the new techniques. The fourth year resident would go learn the new technique and then teach all the attendings during their fifth year, the chief year. And during my fourth year, the new technique was a tep total extra perennial inguil hearty repair and the first in the world to do it was Barry McCurnen and he was in Atlanta. So I spent a day with Barry, watched about four cases and then scrubbed on every inguil hearty during my chief year teaching people how to do it. And as crazy as it sounds, by the time I graduated in 1994, I was already considered a world expert in laparoscopic surgery because I had done so much, not just me, but my co-resident Jeff Tucker, the two residents ahead of us, we are all considered world experts because we had done so many procedures. And I think to your point, the reason why I think I got so interested in trying to measure complications, learn, eventually learn to how to do continuous improvement around the patient process. Again, was because now we are showing all of our complications on a screen. Up until that time, surgery conferences were full of surgeons telling everybody how wonderful their outcomes are, how important it is to learn their technique, 'cause it's the right technique or the best technique. There wasn't a whole lot of the surgeons who got up in conferences and talked about their complications, but that became a really important concept with teaching laparoscopic surgery was to show your complications, to try to help other surgeons avoid what you had been through. And that was something that we did a good bit of because it was so important to share and try to minimize having those same complications as other surgeons went through their learning curve. So eventually I got recruited into academia. It was really interesting, almost every one of the pioneers, and I wasn't a pioneer in laparoscopic surgery. I just had the benefit of the pioneers. They were almost all in private practice. They were not in academia. They just saw that this was better for the patient. It was really a selfless motivation trying to do what's right for the patient. Going from a much larger incision that often led to a long hospital stay, you could do a few small incisions and have the patient go home the same day with less pain than what's really, again, the tender-hearted surgeon wanted to see suffering be less around the concept of surgery. I think for the first few years in the early '90s, they really took a lot of flack from other surgeons who didn't do laparoscopic surgery, especially a lot of academic surgeons 'cause they made them look bad that they weren't doing mental illness surgery and these private practice surgeons were doing it. And that really stuck with me that negative kind of environment that we had in surgery back then, of belittling people and tearing them down for trying to do innovation or new thing. I hope, and I think that's changed a lot in my 30 plus year career. I think there's surgeons and generally, people in healthcare are a lot more accepting of change and innovation. And really it's a brain thing. Don't think that the way we're doing is perfect and can't be better or think we can always do better and we should do better and we should be trying to innovate and do better. I saw that kind of change and I think that prepared me for what we're going through today, which is a total change in how we look at healthcare and science and continuous improvement in data science. And it's about time, I think, since Major League Baseball went through this 25 years ago, I think it's time for healthcare to do a true transformation and apply principles of data science appropriately. 'Cause up until now, we just haven't done that in healthcare and I think that's a major reason why we have an unsustainable system. How do you see data science principles applying to healthcare to transform the system? Do you have any good examples or kind of visionary statements? - First, it was an emery and then got recruited to be Chief of General Surgery at University of Missouri. And before that, I hadn't been Chief of anything in my life, but I felt obligated in a leadership position to study, to try and do a good job. I studied healthcare, I studied business and leadership. And as I studied that, I began to realize that we, in fact, that our global healthcare system is not sustainable. And the reason, the foundational reason why is we're built on this reduction of science concept and it's just not valid. So as I'm studying this, I'm learning this as the Chief of General Surgery, I began to realize we have to do better, we have to improve. And I still didn't understand the science yet, but I didn't want to try to do better. And the first thing I started to do was rebuild the division around patient-centered teams, identifying the patient process from the patient's perspective and then build resources around that process. And so we had a hernia team, a breast disease team, and a GI disease team and started to hire what we call patient care managers, because we didn't have anybody to help the patient and family through the whole process. And it was funny, 'cause the patient care managers, the first two that I hired, one was trained as a flight attendant, one was a social worker, so they didn't have a lot of traditional healthcare experience, but they had a lot of understanding of social issues and team approach. And as they learned their areas, hernia disease and GI disease, and as surgeons, we were teaching them about those things. Within about six months, they started teaching us about what the patient actually goes through in the process of care, 'cause there's a lot of things the patient doesn't tell us as surgeons are afraid to tell us or they're intimidated. And you know, as a patient care manager started telling us, the patients were afraid of this or they had this problem and they wouldn't tell us and we were like, "Yo," and I'm like, "Yeah." And that's that beginning of some of the principles of data science around the diverse perspectives of the small team and how important it is to get different perspectives around a certain process. It's for a few reasons, the leadership changed there, and as I was learning this over five years, this is now around 2009. I really began to learn some of the principles of data science, system science, started bringing engineers in to the patient care and getting their perspective. - Yeah, I think it also speaks to the importance of a multidisciplinary team. Imagine if you never talked to the nurse, right? Like you would never understand how they were trying to provide the best care for the patient and adjusting their chart accordingly. - Exactly. When we function in silos, like we often do, we lose the ability to really learn together. I remember one of the things I did, one of the CQI meetings I did when I was chair of surgery at University of Tennessee, as I brought our vascular surgery team together around vascular access, kidney dialysis, some of the problems they were having in those kinds of patients, there were lots of bad outcomes and complications. And the dialysis nurses came and the surgeons came and we sat in a room and we did the CQI whiteboard session. And although they worked together on the phone like every day, the dialysis nurses and the surgeon never saw each other in person, they were never together in person. And just in that hour we did that day, there were like three new solutions that came up just from communicating together around how to improve patient outcomes. And that was one of the times I was like, "When this kind of method can really be impactful in healthcare when we, to your point, see different perspectives and get different perspectives around the patient process." But I left Missouri and ended up in a community academic medical center in Daytona Beach, Florida. Now I was chair of the general surgery, building a new residency program. And that's where I really had begun to learn some of the deep principles of systems and data sciences as we were doing this in Daytona Beach. I brought in a whole team of data scientists that would work with us every day in the OR, in the clinic. And over a period of years, we learned how to apply these principles to real patient care. And the science is all about measurement and improvement. If you can measure something and use data science tools appropriately, you can improve what you measure. And it took this a few years, but we realized if we're going to have a sustainable healthcare system globally, we're going to have to learn how to measure value of care so that we can, in the context of every definable patient process, so that when we measure financial data with patient outcomes data and use the data tools appropriately, we can predictably lower costs and improve outcomes at the same time. And so we did that in a complex hernia program, where we were seeing patients from all over the country, even we had some patients come from outside the country as far away as Singapore and Australia, because we became known as a hernia center because the work we're doing, a lot of that was because of the continuous quality improvement method that we developed using non-linear tools which can see weighted correlations and what factors are really impacting outcomes the most. And then we can do something about it. And as we did that through multiple feedback, we started getting much better outcomes and much lower costs. And so the hospital benefited greatly financially. You asked about an example of some of the most impactful things that we did. And I think the most impactful thing for me really was identified by Rami, our patient care manager. When we're having a CQI meeting periodically, we'd look at our data, we'd look at our analytics and brainstorm, how can we gain insights from these analyses and apply them? And she was saying one time, you know, these patients are having these bad outcomes or they have these problems and we haven't had great success treating them with surgery. She said, these are the same patients that blow up my email, call me all the time before surgery. They have unrealistic expectations or high anxiety or whatever, it was just she saw a pattern between patients behavior before surgery and their outcomes. And we weren't measuring that at all. So we had to come up with a new measurement. We called it the emotional complexity, high-medium and low. But we eventually started doing for these patients, not every person has a problem like this, but when they do, we started doing prehabilitation and in learning to do things like cognitive behavioral therapy, which is not one thing. There's lots of different ways to do it, but it can actually rewire the brain in a positive way and undo that chronic stress state. And as we learned to do that, we saw that our outcomes got much better. That was a really a hall moment for me. I went four years with almost less than 10,000 year for my salary, which was a painful time for me financially and then fairly painful because the timing wasn't right. There wasn't very good acceptance. Similar to the early days of laparoscopic surgery. But I think healthcare's moving to a value-based care model. What changed to get us to your visionary state now? That was really a long process. It took me years to go through what I think was the right thing, was I kept studying, I kept learning and doing my best to understand. And finally, I knew this had to get into the system somehow. And the hospital in Florida wasn't supportive. I started interviewing around the country to see what was out there. And I had three good options. And the best of the three, I think this is 2015, was I was offered to be chair of surgery at University of Tennessee in Knoxville and a nice place. Great people. I had a former fellow that was one of the attendings there. And so there was some comfort in knowing some people there. So we took our hernia team and our CQI method and we moved to Knoxville. And I thought as chair of surgery, on a leadership position, a really fairly good environment that we would get this done. And for a number of reasons over the three and a half years or so I was chair there, I couldn't get the hospital by and all the way, a lot of issues. The good news is to your point, more and more people were learning, especially clinicians, that the way we're doing is not working. The volume model is not sustainable. There's a lot of stuff out there about burnout and unfortunately even higher than average suicide rate and clinicians. And that's kind of the byproduct of the healthcare system built on an invalid science and pushing the volume model. So the surgeons there were great and they were really open and willing to do this. And then like I said, the hospital wouldn't do it. Eventually, I left in 2019 and during that time, we had with this concept of measuring value and using data science tools for the whole cyclic care to improve patient outcomes, we developed a business model. We learned that if we can measure the value of care for any patient process, we could also measure the value of any drug, device or diagnostic tool in that process. And my two partners, Rami and Brianna joined us while we were in Florida. The three of us said, I guess it's time for us to start this company and run it ourselves. And as we started to do that, the pandemic hit. And I think that was a wake up call for a lot of people in healthcare. I think that was a catalyst to say, hey, essentially, we have a pandemic in 2020 in the 21st century. We're using the same exact science as 400 years ago and it's not good. It's tragically flawed and so many people suffered and died and didn't have to be that way. And I think one thing that was revealed is we don't have a data in information, infrastructure and healthcare to measure and improve patient outcomes. Most of our data management is for documentation, for coding and billing and getting paid, but not measuring outcomes and improving outcomes for patients. And without a data and analytics infrastructure based on data science and healthcare globally, that's why we have such a tragically flawed and unsustainable systems. I think as that happened, it revealed to a lot more people, the unsustainability of healthcare, we began to get some validation of our CQI method. One of our clients wanted to go to the FDA and get a new 510K to remove a contraindication for contaminated fuel. It was a resorbital synthetic mesh. And through our method, we had generated very high quality data, outcomes data and over a period of years with follow up. And we brought that to the FDA and the FDA was very complimentary of our method and the data quality that we generated. And so they gave a new 510K and removed the contraindication for contaminated fuel. We had showed about a third of the patients we had contamination and there was no mesh-related complications, no mesh removal. So it kind of validates. And since then, we've had another FDA approval for one of our clients for off-label use in children, pediatric surgery for low-pressure pneumo-periodium systems. With that kind of validation, it really helped us to show that the science behind our business model is valid and that continues calling improvement and the CQI method that we developed is valid and really better than the way we generate evidence traditionally. You mentioned that the data infrastructure isn't there in healthcare. Do you think that's part of what's daunting about applying data science principles is that you don't have the data and you think you have to measure everything? So how is your approach a little different with CQI and applying data science? Do we have to know everything about the patients or are there specific things you can look at that can really lead to some of those insights and optimizations in patient care? - Yeah, you bring up a couple of really good points. The way data is done in healthcare is based on reductionism and what happens with increasing complexity in a reductionism model, you get more and more fragmentation. And so that's what we see now. I think there's data prisons, little data jails and data prisons, all throughout healthcare work. We've got a data repository in the clinic. We got another one in the hospital. We've got another one in radiology. We've got another one in the OR, the ED. All these data repositories are getting data put in and the data gets trapped. And it's hard to get out and it's hard to analyze. And because the systems are so poorly designed and in the fragmentation of how healthcare is done today, it's overwhelming for the documentation, the nurse, the doctor, whoever's entering the data, they're just trying to get through the day. And so you get a lot of, you know, nobody's fault, but the design of the system and the reductionist structure, you get a bunch of errors, you get a bunch of copy and paste. And so the data is really unfortunately trapped. So that's one point you bring up. We have to free up that data and put it in context of the whole patient process. If there was a design expert who understood systems and data science, we would be designing electronic metrics around the patient process, not into the fragments of care. And so when we do that, this is another principle of data science you bring up. We don't need all of the data. We collect way too much data and healthcare. And almost all of it has little or no impact on patient outcomes. We spent almost all our resources on documentation collecting data. And then we realized we should only be collecting what matters, what are the important factors that the patient brings into the process? What are the important factors on the treatment process? And how do we best measure outcomes that measure value for this in the context of this specific patient process? Because how you measure value will be different for different contexts. You don't measure value for heartache care the way you measure value for breast cancer. And that's one of the problems with our top down approach, trying to push health care in our direction is you can't measure value the same way, either in context or in each local environment. Another principle of data science is decentralization of the data and continuous improvement of what we measure and how we measure. And then there's so many errors and flaws we have to work through to make it a high quality data set. That's another principle that's really important is the way we're doing data today in health care is garbage and garbage out. There's so many errors and we have to work through and resolve data conflicts, data gaps, data errors, data anomalies and we learn to do that over almost 15 years. Doing this now we're working with the tech team to develop automated data acquisition and data cleaning that's gonna really allow this to scale. And so freeing up the data, doing data that matters in context in each local environment and then learning, it's really teaching data science to clinicians. Learning how to improve measurement. We're all measuring the same thing the same way and different perspectives can help improve our measurement. One example, we had been measuring outcomes for complex abdominal wall reconstruction or ventral hernia is the same way everybody does in terms of wound infection because the CDC mandates reporting and how you measure wound infection is superficial, deep and organ space. But I remember at one point we asked some patients who had wound infections after big abdominal wall reconstruction. We said, "What do you think of these measurements?" And they were like, "That doesn't make any sense." And we basically said, "How should we be measuring wound infection?" You've had a wound infection. They said, "How invasive is the treatment required to heal my wound and how long did it take to heal my wound?" We were like, "Wow." That seems like a much better measurement in terms of value than the CDC measurement. But we were measuring short term outcomes because we were looking at a new drug that was a long acting local anesthetic. Most local anesthetics last a few hours, but this new drug could last days. So it's used for post-op pain control. You can imagine a very painful surgical incision. It's numb for three, four days. How nice that would be for the patient. So we were doing the CQI project and we really wanted to be able to measure pain and recovery short term to see the value. And so we decided to measure pain by a visual analog scale, zero to 10 because that was real simple. It's a number. It's easy to analyze. And the nurse has always recorded. And so it's there in the medical records, easy to acquire that. But the first time when we were doing the first analysis, after we collected the data, we saw these anomalies. We saw pain scores were like zero, one, and then eight. And then another patient, there were one, two, one, seven. And we were like, this looks like it's effective. It looks like the drug's working. But then there's this anomaly. And it's not working anymore. It should still be working. And we couldn't figure out the anomalies. We finally went to the nurses who were recording those numbers. And we said, hey, can you help us understand why this number is so high in these patients when it should be low? And it basically says, oh, yeah, we have to lie. We have to put a false number there because the administrators of the hospital put a rule in place that we can only give the patient two pain pills if the score is six or higher. And when a patient may not have any pain, but when they get up and walk after surgery, which is a good thing, we encourage them to get up and walk, they think they're going to be in pain. And so they want to get two pain pills. And so the drug did care for a patient. We have to lie in the chart. And we were like, oh, OK. So we can't use this number as accurate data. So we had to change how we were measuring pain postoperatively. We started measuring in terms of the actual opioids used by the patient in terms of morphine equivalence. Now it's much more accurate, better. God, so many, many examples in health care of, you know, we needed the measuring what matters. And we need to understand how to clean data and not use garbage in garbage out. And that's, as we learn from the data science, it's all science, again, is all about measurement and improvement. Not only can you improve what you measure, but you can improve how you measure. And, you know, as we continue to do that and teach clinicians, how to do that, we'll get better and better outcomes. And again, very similar to what Major League Baseball went through 25 years ago. A lot of people heard of or seen the movie Moneyball. When you read the book, you get into some of more details. And when I asked Billy Dean, what did it take to adopt data science? And he said it took a certain kind of humility. Really had to kind of humble yourself. Because when you first, it's just like I fell in health care. When you first get approached with the idea that we can always do better and you need to change. The first response is, no, we don't. We're good. We're experts. We know everything. We know the right decision given any baseball situation. And he said they really had to humble themselves and realized they didn't know. And with the principles of data science, they were able to do much better. Now, baseball is a little different, right? They're all competing. So once everybody learned it, now nobody had advantage. And it didn't really help that much. But in health care, it's very different. We're not competing when we share knowledge and we share algorithms with each other in context. When we do that globally in health care, we'll be able to have the best predictive algorithms possible that will allow us to match the right patient sub-populations with the best value treatment and identify treatments that could be harmful for different sub-populations. And it's in the full application of data science that I think we can have a remarkable improvement in the sustainability of health care and lower cost that our outcomes in. What that means is we all need each other. And it's a very collaborative science that if we put up walls and silos and we don't share knowledge and algorithms, we'll plateau according to the science. We won't continue to get better. But when we share ideas and we share innovations and we share algorithms, when we network those algorithms, then we will continuously improve value-based outcomes globally for all of health care. I want to touch on what it means to have things that don't work in local environment. What do you mean by that? Is it that a treatment in Birmingham, Alabama? It may not be a, or the right treatment in Birmingham may not be the right treatment in Knoxville or LA, can you tell me a little bit more about that? Yeah, the principle, and again, this is a data science principle, is each local environment will have its own unique variables that are not controllable. One, the simple example is a surgeon's skill, right? If I have never done a robotic operation, I can't apply a robotic approach to this patient population. And there are many other variables that are unique to that local environment. Maybe the patient population might be the devices available through a hospital materials management. You just don't have this device in this hospital. Yeah, think about a surgeon who's in a refugee camp with a surgery center built in a refugee camp that has its own unique local environmental issues. But that surgery center in a refugee camp will have innovations and ideas and algorithms that will be needed by others around the world because that's how you reflect all minorities and marginalized subpopulations. You get those local generated algorithms. What we're doing today in health care is we're aggregating data and making algorithms out of big data and what that produces is averages, not insights. And so there's dozens of articles now published about performance of algorithms. They're generated from aggregated data and they're performing extremely poorly as would be predicted if you know the principles of data science. And this is tragic because this isn't intentional, but it's going against the principles of data science. So aggregating data producing averages is very harmful and it's harming mostly minorities and marginalized subpopulation because they're not reflected by the average. And so we really need to have people learn and understand principles of data science because it's a very robust science now. It's over 100 years old and it's going to be predictable that when we use these aggregated data generated algorithms we're not going to be helpful and they are going to be harmful and that's tragic. Thanks for joining us for another episode of Data Nerds in the OR. A surgeon's journey toward value-based care. You'll find links in the show notes to any resources mentioned in today's show. If you're enjoying our podcast, please subscribe so you never miss an episode. And if you want more content like this, you can always sign up for our smart surgery blog via the link in the show notes. Or if you want to ask Dr. Remshaw or the team a question directly, please send an email to [email protected]. This episode is brought to you by Care Syntax, the leading vendor-neutral surgical intelligence platform. The Care Syntax platform can ingest and analyze data throughout the surgical workflow, including clinical, operational, financial, and outcomes data to capture a complete picture of the surgical pathway. That data can produce actionable insights to enhance efficiency and innovation for surgical teams, hospital administrators, and MedTech developers, driving innovation for the future of healthcare delivery. Care Syntax impacts over 3,000 ORs and more than 3 million annual procedures across the globe. Learn more at www.caresyntax.com.