Transforming Healthcare With Data: AI, Gender Bias and Patient Outcomes with Shirin Towfigh, MD, FACS

lot of new studies that show that women don't present with heart attack pain the same way men do. So, men come in, they have left side gripping test pain that radius or left arm. Women may have it ready to their ear and it's not the same and then the women therefore die of a heart attack and the men get taken straight to cardiologist. You're listening to data nerds in the OR, a surgeon's journey toward value-based care. The podcast focused on data and data science as the keys to a better health care system. Ideas brought to life by the vision and experience of host Dr. Bruce Remsha. Each week Dr. Remsha sits down with different players in the health care 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 Remsha. So, thank you. It's a pleasure today to be talking with a good friend of mine, Dr. Sharon Tofi. Sharon comes with a ton of experience doing podcast herself and she is a hernia nerd like me. We work together in the hernia society for quite a while. We have a lot of similar passions, but she's very well regarded. She's very well published and I'd like to let her introduce herself a little bit more. Tell us about you and then after you introduce yourself, I want to hear about your thoughts on the state of health care data in our health care system today. Thanks so much, Bruce. Yeah, we're good friends and we met in the hernia world. Bruce, of course, is one of the major leaders in the hernia world and a lot of what I've learned has come from what Bruce has done clinically. But we do share the same philosophy. We're very patient centric in the way that we evaluate patients and we are not shy to take on puzzles and difficult medical scenarios. And so we've shared a lot of patients that way. I really love that. We continue to collaborate. So, thanks very much, Bruce. I'm happy to be here. Yeah, thank you. Maybe tell us a little bit and you have a wonderful private practice. You've really not only excellent surgeon, you're an excellent businesswoman. I know I've spent some time with you in Beverly Hills and I know you have a wonderful team. But maybe talk about not how well your practice does, but the overall state of health care and health care data. What's your view on the current state of health care data in our health care system? Yeah, that's a very interesting topic because data comes from all different avenues. I have my own personal database, right? That is data that I control. I know details from every single patient. So it's very granular in the data that I can provide. I also submit my data into a national quality collaborative where the data is pretty much anonymized once it gets to the database and it gives you an opportunity to learn about what happens on a national basis. I think there's maybe four or five hundred eight hundred and that range under a thousand surgeons that are involved in this national hernia database. That's a different database, right? So there are specific data points that are captured in specific data points that are definitely not captured. So it has talked to the minuses. You can make generalizations about clinical data, but you can't really hone in on a specific problem with accuracy. The hospital had his own database, right? So a friend of mine recently who's a surgeon has moved into the administrative world and is a chief medical officer at a major hospital and the data she's seeing is totally different than the data that clinicians see. They look at outcomes, they've got a lot of kind of population data from the hospital from infections and readmissions and use of resources. Most of it is never shared with the physician, but now she's seeing it having had that experience as a physician. So just came back from the European Hernia Society meeting Sweden, Germany, Denmark. They have very robust national databases. They're able to provide 15 year data on recurrences of hernias, for example, again similar to our own national database. It's limited as to the data that they can find, but because people in Sweden tend not to leave Sweden, people in Denmark tend to live there forever. You know, once you're born, you're in this database until forever. So the type of data that we can get from some of these larger national population databases is very, very unique and great. And I'm glad someone like you Bruce is involved because you understand all that. And I think what we'll be discussing more is how important it is to know where the data comes from and understand the population databases different than an individual patient and what happens to them. Yeah, I think you mentioned it. We do a lot of work in various ways. He suggested either at the hospital level or in registry level or a national level. It tends to be only centralizing the data, putting aggregated data together. And certainly that can give you trends in high level averages. But I think what we've not done well is used data in each other's environment across the whole patient's cyclic care. There's lots of gaps. I'm sure the data that you're collecting in your practice is much better than what most clinicians have because you have the resources and the ability and like you said, the control to do that. And that's, yeah, as you know, that's what I did for years with teams of data scientists to learn how to do data science really well. And I think we need to teach that more and more because like you said, it's fragmented. The hospital has this that centralized registries have that in the clinicians. We don't have the data that we need typically to really understand. We're doing our best with what we have available. But, you know, one of the big gaps, as you know, is follow-up data. We don't have any follow-up data. And the other is lack of understanding of the differences in different local environments. That's aggregated centralized data gives you averages. But when you look at each local environment, they're all unique. So you told me about an experience you had working on a project with, I think, with Southwest surgical Congress about SSI bundles. And this was probably much earlier in my learning about system science and data science. But when you describe the project in the surprising results, it definitely clicked that, oh my gosh, that is data science. You can't do the same thing in a local environment and have the same results. Maybe can you share that experience? Yeah. So when I was employed at Cedar Siner Hospital, we were among, I think, nine other hospitals in the nation that were chosen by the Joint Commission Center for Transforming Healthcare to come up with a evidence-based bundle of things to do to reduce surgical site infections in colorectal surgery. And we won the award for Best Project, which was cool. But I worked with these data systems experts and people that were black belts and all these different things that I'd even learned about before. But we came up with a bundle and it involved everything from preoperative preparation of the patient, cleaning them and washing them with certain corhexidine-type props. And then antibiotic dosing and then changing instruments for closure after you touched the bowel, which is considered contaminated, and then changing your gowns and gloves often and having a whole new set of instruments to close the skin after you use the same similar instruments or cutting the intestine, for example, and so on. We show that you can dramatically reduce, I think we went for like 25 percent, approximately, in fact, for eight down to, I think, seven percent is a dramatic reduction. But even within our group of surgeons, we had one surgeon that had a zero percent surgical site infection rate before we started the bundle. So we said, okay, we're not going to change you. Right. I make you better. The left learned from you what you do. He's a lot of bayon and he's just a very good surgeon. Probably his surgical technique was one of the reasons why there was very little tissue injury in there for infection. And then everyone else we mandated this bundle. And then we present our bundle and the other non-hospitals, it included things like Cleveland Clinic and Cleveland Clinic, I think, NYU, bunch of very just large hospital groups that had colorectal surgeons very actively involved. And everyone had their own kind of bundle and some people showed that using their bundle did make a difference. We showed a dramatic difference. And then right after we did ours, there was a publication that followed a very similar protocol and the publication was it makes no difference. Even if you do all that, there's no decrease in the surgical site infections. And you throw up your hands as to what works where and how and how much surgical technique has to do and individual surgeon influence has to do with implementing these bundles. And it's a little frustrating, right? Because there's no answer. And I think as surgeons and doctors in general, we understand nothing is black and white. Yeah. But patients don't understand that often. Like, why can't you just reduce the rate? There must be a couple things you can do to reduce the infectious rate. Why did I get an infection? For example, or why did I have a hernia recurrence, for example? And we can say, well, you have an x percent risk of getting an infection. They'll see, you tell me how we had, I'm going to make this up, a 5 percent risk of infection, why did I get an infection? Well, you're in that 5 percent. Right. To them, it's almost 0 risk, but to a scientist, it's like a real risk. It's just not a very highly expected risk. It's frustrating that we don't have perfect systems. Right. That's why you belong. You'd be simple if we're just a static mechanical system. And we could just figure out what's the right one answer for the entire static mechanical system that's the same. But we're not. We're complex, biologic organisms constantly change. And when you shared that experience, it was one of the aha moments for me that kept me understanding that we are complex systems constantly changing and that we can't do one size fits all. Yeah. And the data science is really the key. Once we have a data and analytics infrastructure, we'll be able to generate algorithms and identify different clusters of patients. Because if you do the same thing for every patient, you got different results. But if you can identify the cluster of patients that could get ABC to drive lower infection. And then this other cluster gets XYZ to drive lower infection. Now you're matching the right patient cluster with the right best value treatment. And other industries have started to do that like net influx and other industries. They look for different patient clusters. The consumer food industry has done this. That's why we have different types of spaghetti sauce, chunky spice or yeah, it's to match with the bet right clusters. Yeah, that was a good story for me to understand that experience you had. Well, I always share your story, which is when you had your lab and you were testing mesh samples, when mesh was being removed, the same exact mesh, sometimes by the same surgeon implanted in two different patients, how to different outcome and how to explain that. That was what started blowing my mind when we did that was just so different than what I was expecting. Because like a lot of us back then, we were teaching it was all about technique. If you put the mesh in with good technique, you're going to have a good outcome. Now that's not always true. Because there are other factors that we didn't know about. That was very helpful. Another topic, there's three or four topics I want to cover. And they all are somewhat related to data, but there are also things that we're both passionate about. And you've done a lot of research in these areas. One is the concept of gender discrimination, gender bias. I know in healthcare, like many other industries over decades, we've seen significant discrimination and specifically in herniated disease. We have some biases that I know you'd have studied very thoroughly. So I wonder if you could talk about in both either or some of the discrimination you've seen, gender discrimination, healthcare, and then also specific to gender bias around data and hernia and things that you've researched. That is something that I've grown to absolutely love about what I do. Back in the day, it's now been 22 years that I've been in practice. So my first hernia society meeting was back in 2002. And I was one of, I think, two female in the entire audience. And the first female to be invited to be an editor for the journal, which is great. They specifically asked me to come on because they've never had a female, but it just shows how you can be in a club and then not really understand that the club is very exclusive. So I think because I was female, there weren't as many females back then who overtly did hernia surgery. I started getting more women to see me and refer to me. And then I would talk with gynecologists and they learned a lot of women, they're gynecologists and all their patients are females. So I started the point where I pretty much have almost 50/50 male to female ratio in my patients that I see, which I think it's about, it should be about a 7 to 10 x difference between males and females. So there's, you should have 7 to 10 times more males that females in your practice. The fact that mine's about 50/50 implies I'm getting a very disproportionate number of women, and they also apply that more women actually have hernias than it is honorably appreciated, right? So yes, I started seeing a pattern, right? They started talking about symptoms that men weren't telling me about, like, pain with sexual function. They were having like a lot of like frequency and urination, worse pain during their periods and so on. And those were all things that we never learned as symptoms associated with hernias in males. And so I've started hearing the same symptoms over and over again and pretty soon, learn that males and females normally have a different predisposition to hernias. But, and we know their anatomy is different, right? The anatomy of the pelvis is very different. The males, the shape is different, the size is different, the nervous system, like nerves that are in the pelvis is much more in women. But also what they complain about is different. So if you're not taught in medical school that a hernias hurt and b, you don't have to have a eutronia sticking out for it to be a hernia and see women can get hernias that it's likely you're not going to be diagnosing these women. And so we started coming up with some talks and I remember we give some talks about these and there's a lot of people, a lot of less of your populations are in, you know, you're in California or you're in Beverly Hills and as Bruce, you and I most of her patients don't come from our pathinity. They travel this yet. So that's not really like my patient population is not all like movie stars and Hollywood people. People like come from all over. I didn't like that kind of poo pooing of what I was saying initially. But now we've gone to the point where people are listening more and I always say that hernias, occult hernias are like chest pain in that gender disparity and you may know a lot of new studies that show that women don't present with heart attack pain the same way men do. So men come in, they have left sight gripping chest pain that radius or left arm. Women may have it ready to their ear or their right arm or they may feel heartburn or anxiety and it's not the same and then the women therefore die of a heart attack and the men get taken straight to cardiologist or cardiac catheterization. So once someone figured out that women present differently, then now we can improve their outcomes from heart attacks in women. So I feel like hernias are similar, not as dramatic. We're not necessarily having more women die. Though you can't argue maybe more women are dying because thermal hernias are being missed and that's the real hernia that has a once missed and if you get to an emergency state, there's a 5% death rate. No other hernia has such an eye death rate. But they're being hurt hurt H-U-R-T, right? They're being hurt. And we showed our study which we presented recently at Pacific Historical Association. Hopefully we'll be published, which is women get about extra over one year delay in diagnosis. They usually present with more pain. They have a significantly better outcome after the repair and they're more likely to be prescribed narcotics before they see me, which is horrible, right? Because you're exposing women to opioids unnecessarily for a completely treatable hernia diagnosis. And many of them have other operations done. They have two ovaries taken out and gallbladder taken out, appendix taken out, they're told they're head. So a lot of delay and unnecessary treatments done before hernia. So the more we understand and talk about that and the more the medical society accepts that and doesn't say, "Oh, that's just "tofies unique patient population, then we're better at it." And I must say at the last several years, we have had specific female hernia topics discussed in all the societies I've been to, including stages, which we had a hernia session. And this year at stages, they had a female patient of actually my patient, but they had a patient as part of the panel to express their frustration. Please listen to us with our public pain. I self-bund a annual research award to anyone who includes gender in their pernea research project, again, to help promote more research into this. And lastly, I would like to say that all that we've done so far surgically, whether it's the mesh we use or the surgical technique or the clinical trials have been male centric. And I have a great slide. It shows, I think, seven different clinical trials, thousands of patients, 17 females, among those patients. Almost all of them had zero females, one of them had seven, one of them had ten. So that's just not appropriate, right? And I would argue the mesh we're using is made for the male pelvis and there are shapes and sizes that should be changed to better treat women's hernias. And then maybe therefore we won't have as many women with that outcomes from hernias compared to men. Yeah, I think it brings up that data science principle that when you aggregate data, it reflects the dominant average in this case, males. And when you do that, unintentionally, you are marginalized populations and minorities in this case, women. And we should learn from that. We should learn from the systems and data science principles where we should equally study and analyze. And when we do data science appropriately, we'll see patterns that are different. And then we can have optimal variety, like you said. It would be a different shape and size of mesh for women pelvis rather than trying to do one size fits all. The data science is really clear, it's agnostic. And when you do it right, decentralized in each local environment, then you can have each local environment reflect the unique variables and minorities as long as you don't aggregate. And then it'll be dominated by old white guys again. I have a question for you, Bruce, because the way that I see it is, and you and I have had very similar practices. We tend not to see the typical simple 30, 40-year-old male with animal hernias. We see the one who's at complications, and it could be like a reaction to the mesh. It could be adverse outcomes with something when we start seeing patterns. So we start seeing patterns in our little niche that the rest of the world has not yet come to appreciate. And you can talk about it. But after many years, eventually the rest of the world comes to appreciate what we've noticed a decade ago. How can that be better optimized? Yeah, I think the honest answer is having a, in all I'll give an analogy to the pandemic, because I think a lot of people are aware that the pandemic was very complex. We had a data and analytics infrastructure appropriately deployed in every local environment. We would be able to see those patterns and generate those predictive algorithms very quickly. When that pandemic hit, one of the problems was it was very complex virus. In some people, it was a horrible disease. And they ended up on a ventilator and died. Other people were asymptomatic. And some people had GI symptoms. And some people had respiratory symptoms. And so the same exact thing like hernia or a virus and different people could have very different symptoms. And if you don't have a data, no one, no one doctor or even a group of doctors is going to be able to figure that out. It's too complex. But if you do collect the data and look for the data point and you described several data points in women with hernia, pain during sex, you know, other types of symptoms that were unique. If we had a data and analytics infrastructure appropriately in place for measuring patient outcomes and learning from them, we could identify those pretty quickly. But we don't. And because we don't, we can talk about it. And we can talk about our own experience with it. We made some progress, but nowhere near where we could if we had that data and analytics infrastructure in place. We're starting to get more. The research is getting a little more along these lines. But again, it's more traditional research than every, it needs to be every patient in a low environment. And so they are starting to see diabetes is not just diabetes. OBC is not just obesity. There are subpopulations and clusters that are different. And that's getting a little bit in that direction. But utilizing traditional research mechanisms is so slow and so costly. We really need to be doing this in real world every day settings for every single patient. When we do that, then the ability to learn and improve will accelerate quite a bit. It's like cholesterol. First, okay, high cholesterol is bad. Like we have good cholesterol and bad cholesterol. Now within the bad cholesterol, there's good bad cholesterol and bad cholesterol. It's just so complicated, but at least we're getting there. I think we are. I think the pandemic opened a lot of people's eyes that because we didn't have a data and analytics infrastructure and healthcare, we essentially did the same thing and had the same result as the bubonic plague hundreds of years ago. We had a shelter in place and PPEs and it's really sad we're utilizing the same science today that we were 400 years ago. And there is a very robust science around systems and data that we need to be utilizing. It's been applied to climate. It's been applied to baseball. It's been applied to entertainment. It's probably time that we should apply it in healthcare. So why do you think it's not a part of healthcare? Is there money issue or an interest issue? They just need leaders like you to. I think it's a mindset thing. We're so trained into reductionism and reduction in science that we think that's how healthcare has to work. And it's a little bit of a fear driven thing because when you understand the degree of complexity and how messed up things are and how much tragic harm happens like we're talking about just in RNA disease, we see it. It's overwhelming. And it took me about an over a decade to learn it. But I think the timing is getting better and better. I think more and more people are recognized in the way we're doing healthcare and the reductionist approach is not working. It's not a solution. So I think we're going to see this happen over the next several years. And to me, I'm very hopeful because if we didn't have this other science to use, it'd be depressing, but there is hope. And that kind of leads me to, I know you've done some work in what's considered artificial intelligence. I know you've done some research. I was wondering because it really is, I think we're in its infancy, right? In healthcare, like you're just talking about, but maybe talk about your perspective on AI, artificial intelligence and what you've seen so far in healthcare and what you see the potential is. The potential is a lot. The research that I did is based on my experience. We have a questionnaire that everyone fills out, we call it the Hernia Health questionnaire. And I've been using it since 2008. So we have thousands of patients that have had this questionnaire filled out. And it got to the point where my nurse usually takes that questionnaire with a patient. And she comes out and reports to me and she'll tell me if the patient has a hernia or not. She already knows based on the boxes that are checked. One of my residents said, really, you can't, you feel like this is predictive. Your question, like it totally is, if I just look at it, I can, I know if they have a hernia or not, because they'll come with a groin painting, let's say. And he can see if we can cope with the score or sometimes the scoring system. So we have what's called hernia scores based on this. And we came up with an algorithm we work with this PhD from Berkeley. You see Berkeley is like, like all into machine learning. And they came up with the algorithms. So you basically fill out the questionnaire and it spits out a percentage risk that your groin pain will be cured by a hernia repair, as opposed to, let's say, a hip problem or ovarian cyst or other reasons to have groin pain. So we thought this would be really great. It was supposed to be launched back in December or now in May. And we haven't launched yet because we then start prospectively using it on random patients. And it turned out, yeah, globally it works, but in individual patients, it's a little bit inaccurate. And so we're adding more data to it to help strengthen that algorithm and make it more reliable because you don't want to put anything on the public that's not reliable. But it's really great because people with groin pain can have, it can be like LeBron James where he's got to add Dr. Ter. Like he's a lady with an ectopic pregnancy. It could be someone who messed up their hip. He could have appendicitis or it can be a hernia. How do you tell who's who? Especially if it's not an obvious hernia or maybe it is not obvious for you, but that's not the reason for the pain you undergo hernia surgery for no reason. So it's really helpful for doctors and also for patients who feel like they're not being heard, they at least have some tool. So we're very excited about our hernia score. We're just not ready to go public with it yet because it's not a perfect system. I don't want to mislead patients by having the wrong score, but it's machine learning, which is a type of artificial intelligence. Yeah, you're describing the application that you can do in your local environment where you collect data, you identify data points that have an impact on what you're measuring. You apply that and then you learn from it and then you update it with more data. You're doing that feedback loop that you need to do to have algorithms get better. I think a lot of healthcare, we have static algorithms and they're going to get worse so they don't form very well to start with, especially if they're aggregated data. But what you described in your local environment to do, my understanding of the science is if we did that in many different local environments and then network those algorithms that can give us the highest predictability. So that's what we're driving towards. Because then you can reflect differences in different local environments and you'd have yours reflects a lot of across the country, but the highest complexity patients and where some maybe a general surgeon in Mississippi would reflect that population locally. And then when you network all those algorithms, you then can have a much higher predictability. That's great. It's like when you get your credit score, there's three different companies you can go through. They all give you different scores. Yeah, it's different algorithm. I'm going to talk about it. We're going to do a session with the data scientists and one of the papers I'm going to talk with her about is there's a paper that is out there that was one data set and they had 29 expert data science teams do analysis and they came up to 29 different results. Because it's not a perfect science and the data the analysis should be used to gain insights not to prove something is 100% perfect. When we talk about gender, a bias, your passion like mine around focusing on the patient and having the patient be a part of the process rather than just doing stuff to them to be talking about about your experience of patient centered care. How important is sometimes they come in very knowledgeable. Sometimes they come in and say do whatever you want doc and but I think it's it's extremely important to understand their fears, their goals, what subpopulations they fit and what clusters they fit in. Maybe share your experience because I know you're very passionate and very good about including the patient in the process. Yeah, absolutely way first job was at a county hospital, the largest county hospital in the US and our system was not very patient-centric. The patient had to conform to whatever resources were available to them. The clinics were, you know, they would show up like 300 patients would fill up and then they would just wait until their turn. There was no orderly said and I was in charge of redesigning all the outpatient clinics. We had something like 700 different outpatient clinics in the LA community system and make it more patient-centric. So that's what I would like my initial kind of introduction to patient flow and making sure the system is optimized for the patient's needs not necessarily the doctor's needs. So that was great and then I moved on to my next job which is at a private hospital. It's a very well-known Cedar Sinai. It was ranked number one in California, number two in the nation. Very patient-centric hospital I learned from other colleagues of mine how to really make sure that you have good relations with referring doctors and and call patients and so on. We knew doctors by their first names as opposed to like anesthesia or GI like personal care and I found that now patients are flocking to me that needed more attention. They're more complicated. They need someone to listen and then I felt even in the system where I was employed I was limited as to like how much time I could spend with some of these people and how I can move my career in the direction of these complex frontiers which is why I'm now my own boss and I can spend an hour or even and rarely but sometimes up to two hours with a patient going through everything that's been done to the figuring out if they want to cry my office we have extra soft tissue paper that my office manager keeps specifically for that because as patients can cry when they've been through a lot and try and help them as much as I can sometimes it's a hurry issue sometimes enough but I'm still there almost like a you know they call it concierge now because doctors that are concierge really try and meet your needs regardless of what it is as a surgeon we're not called concierge the service that we provide can be optimized so that patients don't fall through the cracks and get appointments eight months from now which is not appropriate and are just shuffled around I'm not that surgeon that says not my problem it's it can be not my problem but let me find you the right doctor and I will call that doctor for you we've done the same together you call my office I'll call your office make sure that records are sent and orders are done and can I give you a really brief story one of the founders actually of the American Herne Society very senior surgeon at my hospital really look up to him he really taught me how to really care for the patient individually and meet their needs at all levels so it may not be in his specialty but doctors but patients come to see him just because they know he's intelligent and will address their needs yesterday we had a celebration party announcement one of Ed's patients who he's treated for 20 years they've been he's known these patients but how many certain do you know and know and treat their patients for 20 years not really a surgical thing to do but to have a 20-year relationship but that's ed for you just donate a hundred million dollars to our department of surgery we're now that Jim LaNore Randall Department of Surgery very uncommon to have a department of surgery named surely based on their very personal relationship with ed Phillips and that's something that you just don't see anymore I strive for that but just not seen anymore where you have a relationship with your doctor for 20 years and enough to have the donate. I think a traditional primary care but things have gotten so complex I think it's two things things have gotten money driven and so complex but what you're describing and what you you strive for what I've tried to and my practice was going away from the volume model and go to a value model that patient and I think I'm seeing finally we're talking about earlier the timing I think is getting better and better people are realizing that the volume model is not sustainable and not everybody yet but I if we can show how to do value put financial data with outcomes data and show tangibly with data science how we can measure and improve value based outcomes I think that's going to help kind of transition healthcare to a sustainable system but that's great tell and I said I let me ask a couple of more personal questions as we're closing this session and thank you so much for your time and all these great stories and experience do you have any movies or books that you've seen or read that's very inspirational it touches on improving our world improving healthcare anything that comes to mind that you'd like to share I'll tell you my brain works I like solving puzzles and so any movie or book that kind of aims at coming up with solutions to puzzle or talking about anything related to puzzles is what I gravitate towards I also love stories so I tend not to read books that are fiction I tend to read biographies and autobiographies I just finished Michelle Obama's book which was great and I'm sorry her second book was given to me as a gift just I learned from people's stories and are the sages podcast that I was hosting we call it sages stories because I just learned people's stories and it's you learned so much about people that you little details you never knew about they're up bringing what got them to do what they're currently doing what makes them tick that's my favorite is anything that has like a personal story to it and I think maybe that's why it's been a lot of time with my patients because I'm like what's your story let's just cutting them off and you ever heard of you ever heard of right it's still insightful I think just to what you described I can understand why we've become close friends because I'm the same way I like non-fiction like learning about people companies how Phil night built night key it took him over a decade before he could get alone without feeling like the criminal even though he's generating like 12 million dollars a year in sales every time the the boat with all the shoes came and he had to go to a bank and try to get alone they wouldn't give it to him in the stories that he had to go through the trials and tribulations he had to go through to to become successful that's just one example but I also when I was a kid I had an uncle who would bring over from I forget what magazine but he would bring logic problems I loved so I think yeah we have for some reason our brain is very similar I really appreciate what you're doing because there aren't that many people that would dedicate their time to I think of someone of our frustrating thing to do which is something you're very passionate about and I understand it but it's very hard to communicate the importance of what you're doing to the lay person patients thank very much into black and white industry is looking at just how they can make money and long term projects and things that can turn the Titanic around speak are very difficult to latch onto so thank you for what you're doing and I'm just doing my thing trying to keep telling the famous story hopefully more people listen about women and the need to improve we need to improve our meshes that are out there so they don't hurt people we need to improve the care that women are given so that they're not delayed in their care and become opioid dependent for zero reason and it's what makes us happy one of the things I think we haven't really answered is how do you really build that data infrastructure to truly do real world evidence and I'd love to get your thoughts on what's needed there that's a really good thing I didn't follow up on that sharing you talked about your own data and how valuable that is and you talked about some of the output you're doing with a predictive tool maybe share with us a little bit about how you've developed your own data capabilities and how that came about and how you manage that because I think ideally like I mentioned we should all have a data and analytics infrastructure but most clinicians don't have those resources so maybe share with us how you've done that I mean I trained at UCLA very academic program it's two years of mandatory research and that really instilled in me a very academic way of approaching everything so whatever I what things I like to do I try and back up with evidence and from the very beginning I started gathering my own data when I was in clinical practice and you have residents that are looking for projects I often have residents want to spend their one year of research that's mandated at our hospital with me because I do have a database and therefore they can use that to come up with questions and help get those answered mentoring is a big thing of what I enjoy doing and so we have pre medical students that sign up full time for one two sometimes three years it's nice to have that kind of continuity so I have young people basically that helped maintain it and it'll win right so they maintain the database so therefore I learned from my patient population they get research projects out of it which helps them for their next career and it works out really well I don't know why others don't in my division I'm the only one that has a real database and it's just I mean I don't know if they feel like they're too busy to do it but it's really priorities I think because it doesn't cost that much well and I think the mindset you expressed multiple times since we started talking today is that you're doing it for improving you're not doing it to try to you know prove anything and that's a mindset change I think we're going through in health care my last question is how do we build that mindset in other physician and medical professionals and caregivers to really start to build that brute force or that ground swell or data science and health care so I always say you can't it'll be great if a resurgence had interest in research or interest in looking at their data but I say take those who want to do it and empower them to do it give them resources, time, money, whatever to do it and then the other surgeons that don't want to do it maybe they're really good at just taking care of patients right have them do that and don't demand that they also do research for example and just have them collaborate so the researching surgeon for example I'll take care of your database right I'll do the database you just keep operating and don't expect the clinical surgeon to also do research or the really good outcomes research surgeon to also maintain a very heavy clinical burden and instead bring them together as a yin and yang so each one helps the other come up with a division now that's very strong and has people to help each other learn that's how I would run a division for example I don't see that done everyone and one has the same kind of you have to do this many surgeries you have to research you have the teach like everyone has to do that and some people don't have strengths in one of those yeah it's like we're talking about we're patients are in clusters because we're human beings doctors are in clusters too right yeah they have people do what they're good at and what they enjoy doing and they empower them in doing that given the resources to excel what they're really good at then they'll be happy and productive okay that was great anything else sharing I can't thank you enough and I hope we can get together a little longer time in person yeah I'm needing I'm definitely going to be at the hernia meeting in Chicago we'll hopefully spend some time together there thanks so much 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 remcha or the team a question directly please send an email to podcast at caresyntax.com 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