Bruce Ramshaw, Caresyntax - Making Surgery Smarter | LSI USA '24

Bruce Ramshaw  0:00  
Hi. I'm Bruce ramshaw,

Chief Medical Informatics Officer at care syntax. And I want to talk to you about our clinical data as a service, and how we're applying data science to real world patient care and partnering with industry clients to help measure improve value for patients, but also assess and highlight the value of your products. I think our biggest asset is the 15 years we spent working with a team of data scientists to learn how to do data in a high quality way. And healthcare. Data is a mess in healthcare. It's in fragmented repositories, it's poor quality. We've learned how to clean the data and create high quality datasets. And I'll talk a little bit about that. Our care syntax platform is agnostic to data, we can ingest any type of data and then apply the data cleaning and analytics to produce high quality out outputs and meaningful insights. The process is a published patent. Again, we spent many years learning how to apply data science to healthcare it starts with a definable surgical process. That's contexts, data requires context to be meaningful. And we identify engaged clinicians working with our industry partner, we define specifically what matters in the patient factors, what matters in the treatment factors, where your products are, and how do we define define value based outcomes in the context of that surgical process for that local environment. And then we do feedback loops to curate the data. To produce that high quality data set and meaningful insights of care syntax, we do the work, our clients don't have to touch the data, we do the data collection, cleaning, analysis, and eventually partner together to produce those value based outcomes. We're differentiated than that. We look at the whole cycle of care, we look at all the data that matters, and that whole cycle of care, we get meaningful data. We also lower the hurdles, our CPI method does not require IRB submission. So we don't have to go that through that process, we get data faster, and so we can charge less for our service compared to traditional clinical trials. And the return on investment is much faster, whether you're looking for return as far as scholarly productivity, publications and presentations, marketing materials, or even influencing an FDA approval for new indication removal contraindications, we can generate that and typically months instead of years. I'll give you one example, we worked with a company called surge request, and 2014. They had a low pressure pneumoperitoneum name technology. Typically when we use a numeral parent name. For minimally invasive abdominal surgery, we have a pressure at 15 millimeters of mercury. That's pretty tight. That's a lot of pressure on the diaphragm. And the abdominal muscles with air seal their low pressure device, we could operate at half that pressure and still have the same exposure. But they were only getting market share in the robotics cases, which was less than 10% of the market. And so we did a seat UI project with them in laparoscopic ventral hernia repair. And nine months we had over 100 cases we demonstrate the value. We wrote a white paper together and that white paper was used by their Salesforce to open doors they couldn't get through before and that led to accelerating market share growth and laparoscopy. Six months later, they were acquired by CONMED for $265 million. And the CEO said our C QI data actually improve their valuation by 10s of millions of dollars. CONMED continues to be one of our clients. And we just did another CQ AI project in pediatric surgery where the FDA approval was for patients who a 20 kilograms and heavier but pediatric surgeons were using an off label because lower pressure smaller kids and just make sense. So we work with Children's Hospital of Philadelphia Children's Hospital in Georgia, we collected over 130 patients half below 20 kilo kilograms half above. And we saw that there was safety and effectiveness. They brought it to the FDA last year and they gave them expanded indication in pediatric surgery, all based on our C QI generate data, no IRB, no clinical trial. And I want to show you a couple of ways that we interact with data with our clients and our clinical sites. This is all the ventral hernia repairs done at one hospital over the period of a year and we looked at patient outcomes like length of stay but also financial outcomes. And we found in this hospital for this year, the average ventral hernia repair was a loss of over $2,000 per patient, but a range of 10,000 profit to over $32,000 loss. And we're looking at one of our clients drugs X Braille it's a long acting local anesthetic. Most local anesthetics last a few hours, extra can last days, so you can match and how nice that is first post surgical pain control for a painful surgery, but extra cost $300. The short acting local anesthetics cost about $1. So the manager the pharmacy budget doesn't want anybody using expro because the negative impact on their budget, but that's not what matters to the patient outcome, or even the outcome for the hospital. So when we looked at the data, when xpera was not used length of stay was over 70 hours, and the hospital lost over $3,000 When expro was used. length of stay was just 36 hours, and the hospital lost about $1,500. So there's a $1,500 net benefit financially, despite the $300 extra cost. I don't know of any hospital that looks at their data this way. We have another client academic medical center in the Midwest, and their surgeons wanted to look at their ventral hernia repairs, they were concerned about the outcomes, especially most recent financial outcomes. So we first looked at RV use. Most hospitals think that if they increase RV use for their surgeons, they'll increase profit. But when we looked at the profitable cases, the RVU actually was less. So there's no correlation with profit by RV is. So then we looked at the actual profit or loss by year and the first five years till 2022 that a pretty steady profit about $700 per case, which translated to about 100 $100,000 net profit. But we looked at 2023 profit tanked to a loss, about $100,000 worsening net profit. So we looked into the data. And we found there was a coding change. They started coding by hernia size. So we looked at the smallest hernias, three centimeters and we saw they're making a profit, but only if they did open surgery, or laparoscopic surgery. If they did robotic surgery on those small hernias, they were losing almost $1,000 per case. So then we looked at the large hernias over 10 centimeters. And we saw they're losing money, but only if it was done open or laparoscopically. When they use the robot on large hernias, they made nearly a $3,000 profit. So we quickly by interacting with the data, we were able to help those surgeons see how they could return to profitability. I just want to thank you. This is really what major league baseball did with Moneyball and was 30 years ago. So we're now finally getting to a point in healthcare where I think it's time to apply data science principles. And when you read the book, Moneyball, and you learn a little deeper about how that happened, Billy Beane, was the general manager, the Oakland A's, and they asked him in the book, what did it take to embrace data science? He said, Well, it took a certain level of humility. We thought we knew what to do in every baseball situation, because we were baseball professionals. And what we learned we really didn't know. And we had to become humble and allow us to interact with the data and really learn what really is the right thing for this situation in this environment in this time, and I think we're ready to begin to do that in healthcare. I want to thank you very much.