From Averages to Individuals: How AI and Embedded Research Are Reshaping Patient Care

Jimmy Bechtel sits down with John Worden, Chief Commercial Officer at Javara, to explore how AI-driven data analysis is moving clinical research and patient care away from one-size-fits-all medicine toward truly personalized treatment pathways. They discuss how embedded research sites inside healthcare systems are uniquely positioned to put these tools to work, what end-to-end data integration means for feasibility, patient identification, and long-term outcomes, and why scaling AI in healthcare is ultimately less of a technology problem and more of a trust and collaboration problem. John also shares a real-world example of how the research process led to a life-changing diagnosis for a patient who almost slipped through the cracks. 

Transcript

Jimmy Bechtel 0:00

Welcome to SCRS talks provided by the Society for Clinical Research Sites. Thank you for joining us as we explore the latest trends, insights, and innovations shaping clinical research today. I'm Jimmy Bechtel. The Chief Site Success Officer with the Society, and I'm joined today by John Worden, the Chief Commercial Officer with Javara. And John is here to share and talk a little bit about, AI and data-driven analysis and, and how we bring. AI driven data into the site operations and enhance, the work that is done at the site as well as patient care. So, John, really excited to jump into this topic. It is a, it's a hot topic as you probably understand and can imagine. It's what everyone's talking about. It's what everyone has question marks around. So, again, cannot emphasize my excitement enough to dive into this in this short 15, or so minutes that we have together. But before we do that. I'd love to hear a little bit more about you and, and maybe a little bit more about Javara.

John Worden 0:57

Yeah, thanks Jimmy and, and thanks for having me today. We're really excited to spend some time on this topic and, and thanks for all the work you guys are doing at SCRS. We've enjoyed our, our eight years of, being a member and, you guys are really getting a great voice to companies like us and other sites, so thank you. I appreciate it. yeah, a little background myself. So I've really been, kind of supporting and serving the life sciences, industry, since the 1990s. I really started in the nineties and early two thousands when life sciences was in the process of moving to automation. They were really implementing technology back then. the initial sales systems, SFA, now called CRM. The initial EDC and CTMS systems were being rolled out back then, and that's really where my career started in the, in the clinical research business. I then, spent 15 years, in the CRO world, and spent a lot of that with INC research who eventually became Syneos Health and held multiple roles there in, in BD and, and operations. And, obviously learned a lot about clinical trial operations with my time at, at Syneos, and then also spent a little bit of time, working at CDMO, learning the manufacturing size of, of the business and, and how it, how, how much work is involved in getting drug to our sites for these trials. And then the last three years I've enjoyed, being the Chief Commercial Officer at Javara Research. you know, we've been on an eight year mission to, to deliver clinical trials at the point of care, and we've really been doing this through partnering with healthcare systems, which we'll talk a little bit about today. I'm really, my career can be summed up as trying to always be a problem solver. I, I'm, I'm really always saying, you know, questioning, can we do this better as an industry, as a function? And I think that leads us kind of into our topic today, right? So, how can you know, AI driven data, advanced diagnostics, help us move, you know, from standardized care to a really more of a truly. Personalized care across both clinical research and everyday patient care. So excited to, kind of dig into that.

Jimmy Bechtel 3:00

Yeah. As am I, John. Exactly right. You know, you're, you're. Really colorful and and varied career, I think probably brings a really interesting perspective into this. It's always interesting to have conversations with individuals who have, been in different areas and, and, and again, bring that important perspective. But I, I do want to get into the, the topic here, and we are really presented with an amazing opportunity. It's, it's scary, it's challenging, it's new. and we as an industry, as you probably, have experienced very directly. Are averse to change. We seem to have a problem with picking up new things, and AI is kind of a beast. I want to talk a little bit about how can AI driven data analysis and these advanced diagnostics that artificial intelligence really enable for us, shift us from standardized care to a personalized treatment pathway.

John Worden 3:53

For a long time, healthcare has been built around averages, right? But there's really no such thing as an average patient. so AI is really finally giving us the ability to treat patients as individuals and connect research and care in a much more meaningful way. And, and what, what is actually changing here, right? Well, historically medicine. Has been very protocol driven. You diagnose a condition, you follow a guideline, and you hope the treatment works for those patients or for most patients. But what we're seeing now is a shift towards really understanding the individual patient profile in a much deeper way. And what's enabling that is ai. Tools like Tempus and Flatiron Health are aggregating clinical data, genomic data, real world data to help us identify patterns across millions of patients and tools like Google DeepMind, they have, they're working in the imaging space and have shown us that AI can detect ophthalmology diseases. Very early on, and even cancer signals very early on, much more than the traditional methods. So what does this really mean to a practice, right? I think in, in practice, we're gonna be able, and we can start to diagnose earlier, which is key. We can stratify patients better and we can start predicting outcomes instead of reacting to them. So instead of asking what is the standard of care We're asking what's the best care for this patient based on everything we know today, and that's gonna be a shift in the way we, we manage patients.

Jimmy Bechtel 5:37

It's really incredible to think because, you know, we couple that with the complication, the, the complexity and the the ness. Necessity to do some of that work in clinical trials as well because of, as you alluded to, the na not only the nature of the condition, right, the nature of the disease, but also the nature of the treatments being used for that disease and these genetic components that are coming into play when we talk about how we approach our treatment to disease. And it's, it's a concept we've known for, for years. We know that not all patients are treated equal and we haven't really been effective. Again, as you've alluded to, in cracking the code with how we, how we approach their care, whether it be, you know, in, in medical practice or in, clinical research as a whole. So it's, it's a really interesting path that we are on now where we have a, a potential opportunity, a tool, and, and, and like you had mentioned, organizations that are going down this path to help enable this and, and provide that specificity that is needed to effectively treat patients in a modern way. So John, then what role can shifting this, you know, more succinctly to our space embedded research sites play in integrating some of these AI tools and this, this data analysis and diagnostics across research and routine patient care.

John Worden 6:58

Javara spends a lot of time, trying to figure out how this plays out inside healthcare systems where patients are being treated every day. And, you know, really what we're seeing is, you know, AI is, is only as powerful as the data that's there and that data has to be clean, but it also needs contextual data and longitudinal data to really understand a patient's full journey. in an embedded care site, when we're embedded in the healthcare system, we really can kind of see a few things that are happening. One, we can see that, you know, actually bringing AI tools directly into the clinical workflows enables a lot of opportunity for us in research. So a patient comes into a clinic. And in the background, for example, an AI platform, something like Tri Genetics or Deep Six ai, they're scanning that patient profile in real time as that patient's being seen by the doctor. And it's looking at things like diagnosis codes, labs, demographics and ultimately it's, it's really telling us that patient potentially is may be eligible for, you know, study A, study B or study C. And this is a completely different model than what historically has been done, which is, you know, a manual chart review and a reactive recruitment model. So that's the first thing that we're seeing is, is already a change that's taking place. Secondly, it creates this really important feedback loop, between research and clinical care. So some of the data that we're using in care can be shared research and vice versa. The research data can be shared into clinical care, and that really creates opportunity for us to create these personalized pathways for treating patients. And third, area that we're seeing is it also helps us really identify and reach patients. That historically have been underrepresented in clinical trials. so those three areas we've seen, as an ABE of health system, being inside the healthcare system is an area that we're seeing a lot of growth. the opportunity really is, we're not just looking at a narrow slice of patients, but we're actually being able to engage in a very large population of patients. And that's being enabled by ai.

Jimmy Bechtel 9:17

It is really incredible, John, and, uh, what a time right? To, uh, to be in research and, and what an opportunity here. We've talked again, we've talked about these things for so long, and we've, you know, we've gone back and forth about what the best solution is and, and some organizations have done a great job at, you know, marginally improving with the progress in those. Three spaces that you've made, and they've had one-off solutions, but nothing's really stuck and nothing has been implementable at scale. So I'm really eager and excited to, you know, see the potential of organizations like vaj, Javara, and the partners that you work with to bring some of that in and provide examples of how we might bring this. bring this at scale and allow and enable other organizations to, find success in this space that we've been unable to attain, previously, particularly in those three areas that you had mentioned.

John Worden 10:10

Yeah, and parti, we just, so Jim, we just saw an example of this in, in the last few weeks where a patient, was interested in a trial. They were part of a healthcare system. They were being managed by that healthcare system. They then said that they wanted to participate in this trial. We did some screening for that patient to see if they were eligible for the, the trial, which included a, a blood panel that was specific, for the protocol. And that blood panel came back with some elevated levels, that obviously disqualified the patient, but we were able to bring those, that detail back to the treating physician, the, the physician that treats that patient. And through additional, work and referrals that patient was diagnosed with uterine cancer. So there's a really great example of how research can really. Yeah, help, an individual have a better, healthier outcome related to, their, their clinical pathway.

Jimmy Bechtel 11:03

It's an excellent example and exciting as an opportunity in and of itself, right? Not necessarily, exciting for that particular case, but as an example, like you had mentioned, John, it, it is really, great opportunity that we're starting to see in a direction that we're headed as an industry and, and, and being able to solve, like, like I had mentioned, some of our toughest challenges, that we've seen over the last few years. So. John, how can then end-to-end data integration across the research and care continuum? Improve areas like feasibility and identification and long-term outcomes.

John Worden 11:39

This is an industry that typically is slow to change and also is pretty fragmented, and that's what we're gonna, that's what we do see with healthcare data, right? It's still very fragmented. You have EHR systems, you have lab systems, you have imaging data platforms. You have wearable devices. Your Apple watches, your order rings, your continuous glucose monitors. And a lot of these don't always talk to each other, right? So, but what, when you start connecting those data sources and you bring that together, this is where AI becomes really exponentially more powerful, is bringing that together. And we're seeing platforms like Snowflake Healthcare, life Sciences Cloud. They're enabling large scale data aggregation. So bringing all of that together. Which now is empowering these AI models to sit on top and to continuously learn about patients. And every time a set of data has changed, they're learning more about you. And every time something on your watch is, is changed, it's learning something more about you. And that's really where we can start to drive real time personalized care for both research and for clinical care. And so what does this kind of enable around the areas that we work in research. First and and easy is better feasibility, right? Instead of guessing whether a site can enroll patients, sponsors can see real world data every day as it's changing and understand exactly. How many eligible patients exist and today might be different than tomorrow, right? That's one opportunity for us in the feasibility and our sponsors. Second is this area of continuous patient identification and instead of a one-time recruitment push for a trial, AI can continuously scan patient populations. Surface new candidates all the time as they come into the system, as their data is updated into the system. And you really have this real time, patient registry or database that is constantly being updated by all these different data sources And that obviously creates an opportunity for patients to know about trials based on their latest data. And it, it also creates an opportunity for our sponsors and sites to identify those patients. And third. Is, I think this is a big one too. I is the opportunity for longer term outcomes, right? Why do we only collect and look at the data for a trial that lasts three years? If that person's in a system, that data's still being collected, on that person, not maybe for that trial, but data's still being collected. And I think there's a huge opportunity for us and the pharmaceutical industry to follow those patients beyond the trial and to see how these therapies perform in the real world, across different populations over time. And that really starts to create a true learning health system. And that's what we're excited to, to see up, come out.

Jimmy Bechtel 14:41

I couldn't agree more, John. It's that last part there is really exciting. It seems that so much of what. You know, there's the primary research, right? You talking phase one through three here and even to some phase four that we seem to always hyperfocus on and we're driving, right? That's, we're trying to get drugs to market here, right? So it makes sense. and treatments and we're trying to cure, and lesson burden on patients, for a variety of different things, but. Where we seem to always have issue with is exactly what you had just mentioned, this long-term safety and efficacy work that we have this opportunity through more intelligent and proactive data collection to be a little bit more nimble, I think, in that process. And not only continue to you know, manage and, and, and monitor the. Treatments that we are producing. In the traditional sense, but also make smarter decisions going forward upstream as a result of what we're seeing here. And, and being able to then iterate as we do as an industry more succinctly and specifically on what's next based on what we've seen and what we've been able to track a little bit more effectively, in a more simple and efficient way

John Worden 15:57

I mean, yeah, I love that because, you know, you're exactly right. What if we find out that, you know, after the trial's over, you know, a certain, population, a certain set of patients with, a certain marker history. Are responding way better on, on a, on, on that drug than originally anticipated. Like that information, you know, unless we go do a phase four trial right now, you know, it's gonna be hard to get, but the data's there. So we should be trying to figure out, now that we have maybe these AI platforms to help us, we should be able to track that and really find out where are the impacts of that medicine and where maybe it's not working as well that we need new solutions.

Jimmy Bechtel 16:35

being able to trust and find reliability in that information because we, you, we as a, as a health broader healthcare industry, make a lot of post trial and unpure reviewed assumptions about medications and treatments and this, that, or the other that's out there. And people, you know, use those to unfortunately push agendas as a result of, of that, Or what have you. And it becomes, a little bit convoluted in sort of, you know, off-label or alternative label, usage of these medications. But we again, will be able to move, I think, a little bit more nimbly, and effectively, right, we'll be able to get to an era. Ideally, this, maybe this is a little bit visionary on our part, but to a place where we have, like you had mentioned, the data exists and we can capture it more effectively. And then, and then, you know. Perpetuate, information and build that into what comes next with more confidence and reliability and, and assuredness going forward. I wanna begin to wrap our conversation up here, John, with, with a final question. How can sponsor Insight networks, in whatever shape or form they might take, collaborate to responsibly scale AI enabled innovations, but also importantly maintaining trust and data integrity, But, you know, kind of your, your, perspective on, on how we, again, can collaborate more effectively to maintain, trust and integrity.

John Worden 18:02

Yeah, I mean it's, you know, AI and healthcare is not, not really a technology problem, or not just a technology problem. It's, it's a trust problem. It's a governance problem, and it's a collaboration problem. So, you know, what can we do to scale this in the right way? And first, and, and always utmost is we need patient trust, right? We need patient trust, and patients need to understand how their data is being used. More importantly, they need to see the benefit. If AI helps them get diagnosed earlier and matches them with treatment faster, that builds trust, right? That makes them more trustworthy of you having ai, helping them with their healthcare. Secondly, this data integrity and quality piece, I mentioned it before, but AI is only good as any technology is only as good as the data behind it. If the data's incomplete, if it's biased or consistent. The outputs from AI will be too. So we need to make sure that our data is high quality and has high level of integrity. Another piece that we need to consider is the ethical components of ai. We have to be very intentional about avoid avoiding bias, especially when we're trying to improve, ongoing diversity in clinical trials. The last area, which, which really is extremely important, is how do we collaborate between the sponsors and site networks to make this happen? And this, this isn't something that sponsors can solve alone. It's not something that healthcare systems can solve alone. It requires a shared infrastructure. It requires all of us to be aligned on incentives, and then it requires really strong partnerships between sponsors, between sites. Between healthcare systems at Javara, we think we can serve as that bridge, right? We can connect the sponsors, we can connect the healthcare systems and the patients, and making sure that these innovations actually translate into better and healthier outcomes for patients.

Jimmy Bechtel 20:00

Well, John, I think that's an excellent place for us to end the conversation. And, and again, thank you for your perspective and, you know, the forward looking insights here. I, I really want our listeners to not only take away some, you know, stronger insights and maybe perspective from the conversation, but also. Areas in which you might look to see advancement and, and really lean into what we can do as an industry and scale that going forward. So again, John, thank you for your sharing your thoughts and your, perspectives and and experiences in this space.

John Worden 20:32

Thanks Jimmy. I really appreciate it, enjoyed the conversation and excited about to see, see where this is gonna go. And excited about the opportunity for, for patients, for, for, for research and for healthcare. So thank you.

Jimmy Bechtel 20:45

Thanks again, John. And for everyone listening, thank you again for your time today and for tuning in. I hope you check out other great site focused resources made available to our entire community on our website, my scrs.org, including other podcasts, webcasts, and other AI related content. Again, thanks for listening, tuning in, and until next time.