Beyond EDC: Designing a Connected Data Strategy for Predictive Clinical Development - IMS HEALTH HOLDINGS INC

Hello, everyone. I'm Jeannie Northrup, joining you today from Applied Clinical Trials. I want to welcome and thank you for joining us today for today's webcast. Today's web cast is entitled Beyond Electronic Data Capture, Designing a Connected Data Strategy for Predictive Clinical Development. Today�s program is sponsored by IQVIA. IQVIO is a leading global provider of advanced analytics, technology solutions, and clinical research services to the life sciences industry, IQVIA creates intelligent connections to deliver powerful insights with speed and agility, enabling customers to accelerate the clinical development and commercialization of innovative medical treatments to improve healthcare outcomes for our patients. With approximately 82,000 employees, IQVA conducts operations in more than 100 countries. You can learn more about the company at www.iqvia.com. Today's program is designed to be interactive. We have a really great panel joining us for today's discussion. They put aside a lot of time to help present to you. They have a great discussion plan. We hope you will take this time to ask them any questions you have while we have them available today for their presentation. You can submit any questions you have throughout using the Q&A box. That box can be found at the bottom of your video player. We look forward to responding to as many questions as time allows of the conclusion of today's event during our Q&As session. If by any chance we run tight on time, don't worry. Please continue to ask any questions you have. Our subject matter experts will actually be fielding and following up with you following today's broadcast as well. For viewing purposes, you can enlarge your slide window by clicking on the small icon. You'll see that located in the bottom right corner of the media player. Please note that all slides will advance automatically throughout today's event. If you happen to experience any technical problems viewing or hearing today's presentation, please click on the question mark help widget. You'll be able to see that in the top right of your presentation window. It is now my pleasure to welcome today's subject matter experts, our great panel. First, let me welcome Sabrina Steffen, Head of Clinical Data Management at IQVIA. Sabrina, welcome. Thanks so much for joining us. Thank you for having me. Absolutely. It's a pleasure. I also want to welcome Jeff Knoll, Vice President of Connected Devices for Patient-Centered Solutions and Connected Technologies, also with IQVia. Jeff, welcome, and thank you for joining me. Thanks. Great to be here. Absolutely. Absolutely. Also, again, I'd like to welcome another IQVIA colleague, Anthony Mikulishek, VP of Commercial Lead Patient Strategy and Corporate Development with IQVia. Anthony, welcome. Thank you, Jeanne. Thank you. And last but certainly not least, I want to welcome and share the floor with Christina Lentz-Larsen, our Senior Director at IQVия. Christina, thanks so much for joining us today. Feel free to go ahead and take the floor. Thanks, Jeannie. And welcome, everybody. So let's start really with the challenge at hand. Why are we here today? Well, we have more data and technology and automation available to us than ever before, but our clinical development timelines really aren't getting any shorter. So where do we see the biggest opportunities to change that problem today? Yeah, Christine, it's a great question, right? And one that we're faced with again and again. I think we've seen the total development timelines for a single asset now increase up to about 10 years, where not that long ago it was eight years. And we see that we are seeing the longer timelines. Um, and, but you would say, because we have all of this technology, this automation, this AI, like what, what is happening? How come we're not seeing it, um, move faster. And I really think that it's, um. It's because we haven't pushed ourselves to truly operate differently. Um, And that we have brought in all of these, this tech and AI, and it is really augmented our process, but we haven' allowed it to transform us just yet. and we haven't looked at kind of that foundation to say, how can we be different with all of these technologies that are available to us? And so I think it's really saying, how does the technology open the door? And it has helped us in some ways, but how do we really transform the way that we operate into a more fundamentally different way of operating? And if we do that, then I think we really start to see the changes in those overall timelines that we're all looking for. Yeah. And I think that we need to start thinking about the different technologies, the automations, AI, and so forth in our operations sooner, because we tend to solve for that after our protocols are developed. And while we're building data collection methods, that we are thinking about, oh, what data do we need and who's going to take care of it and what are we looking for in risk? And really, we need to start bringing that sooner in the process to become more predictive. We need to anticipate what we think needs to go right for the study and then build our processes and data collection methods around that. Yeah. I would jump in here to say that But in my space, as our teams bring a lot of digital health technologies to bear to client strategies, I think what's resonating when I listen to the two of you and your comments are folks look at our group and say, automation, new tech, this should go fast. And when you just think of the practical challenges of stacking up technologies to create an endpoint strategy, it can really do the opposite. It can cause folks to question, it can cause processes to have to be re-evaluated. There's just the practical nature of bringing on different players in this space can slow things down. So to your point, Sabrina, if we are going to start to take more advantage of the technology and actually look at customizing strategies or changing the systemic way about building a trial, that may bring new interesting opportunities to bear. And so I think if you're taking into account data quality, you're taking into count the mechanics of the trial strategy, you're certainly taking into accountable user experience and how are we going to use this data both operationally as well as the back end. Christina, it kind of gets the conversation going around, well, how do we look at it more holistically? And what's the timing of starting to think about strategy? Because it can't all be done at the last minute. It sometimes happens in our business. Yeah, I think so much of how we collect data and what data we need has to come from what's critical to the study. And we need to think About that before the study starts to run. So really, it starts at protocol development. And We need to focus around what's going to put that patient safety at risk or the study results at risk? And then what's the information that's going to help us recognize that those are falling out of control? And really that sets us up nicely to address that risk early. And so doing this strategy upfront during protocol design then lets us think about holistically how we're going to capture the data we need to best collect it and then make it available for decisions faster. And that's where the digital, you know, the digital health, digital first and digital first data collection, digital health technologies, that's where that really comes in. We need to think about this while we're designing the study, not wait until we're ready to start the study. That's the key. Yeah. Yeah. I think there's, let me, if you don't mind, Jeff, I'll just jump in here real quick too. I think the digital first, in the technology is really important, but, you know, the science, we're all in this for the science and science for the patients, but there's also a science behind how we collect that data and the science behind what data is collected, especially when we start taking a look at pros and clinros and the signs behind which pros and clinicals can be, can be can be or should be collected. And that can have an impact on the design that could have an effect on the amount of data that can impact on some of the predictive analytics that you can do during the study. So I think the science is another aspect of this is that we need to keep focused on as well to help have predictive outcomes. Yeah, those are two good lead-ins too. And when some of those conversations start to come downstream then to my groups and folks say, okay, well, what technologies, what medical grade devices can we bring to bear then for that endpoint strategy across a protocol. I think if teams are in this space and they're used to working with medical grade technologies, then they're also used to other things that IQVIA teams and other folks that play in this place bring to bear are intelligent platforms. Intelligent platforms that provide an e-source opportunity to take humans out of the mix. And you can create a data acquisition strategy direct from a patient into a platform, and it doesn't have to be more traditional platforms. It starts to uncover new opportunities to go faster, new opportunities to make different sort of medical review decisions, monitoring decisions, et cetera. But if you're not in that space, I think you continue to get a little hung up on the traditional ways of trial design. And then that brings into account things like regulatory and other types of questions. And so the more used to teams are in working with e-source strategies and creating digital health technology strategies that pull data direct from patient into a platform, they get a little quicker, I think, to see that, oh, there's new opportunities here to do trials differently. So I think if you do start to evolve the view on what trial design looks like, Tony, it doesn't always perhaps have to include EDC at the center of every design. No, well, we start taking a look at the EDC landscape and where we've been over the last couple of decades. You know, it's a great replacement for CRF, and there's a lot of great EDC systems out there that we've all used to help collect that clinical trial data. However, as you've just talked about, all the DHTs and other types of components that are out there and that are being used, there's about 30% of current data stored in the EDC for clinical trials. We have all these other systems that we have to work with and all these end-point data that we Have to manage. So it's actually going beyond the EDC right now. And then, so when you take a look at that, you take look at all these other systems that we're working with and dealing with, and we've all done this, you have to start not connecting all these systems together. And we talked about doing things quicker, doing things faster. We've been in the industry, some of us have been in industry for decades, since the day one that I've been in the history of always tried to do things quicker and faster. It's still, that's the case. by adding more technology, by adding integrations, by adding reconciliations, are we really making trials faster in connecting all these things together? So I think it adds a layer of complexity. And so just an EDC alone is not the best way. And what we're actually creating is a federated environment of clinical data. So definitely it's adding to the complexity of all the massive data that's being collected. Yeah. Yeah. And I think teams that are a bit more digital first in their thought processes, they start to get experience in the way that these stacks of platforms work together, if you will. And from ePro, Clinro, through medical devices, on the operational side of the trial, all the way through to the back end on making the decisions and what the data management and stats teams need for the data collection. We are driving it forward, but if you're going to move the needle and you are going to actually automate and go faster, you do need to take a look at the foundational parts of your trial strategy. Yeah. I mean, I completely agree with you. You know, it's really interesting when you take a look at EDC overall. And you think about like, you know, I go back, I started when we were still using paper and went through that transition from paper to EDC. And we all thought EDC was just going to solve all of the challenges and it was going to take care of everything. But in reality, I see EDC as kind of being that first digital data collection method that we brought into place. And since then, you've seen, of course, we've had central labs in place for a long time, but you've seeing the increase of ECOA and connected devices and other digital data collection methods. And so I say, now I kind of look at it and say, EDC just kind of paved the way for that a bit more. But I think now we're in this place where the other digital data collection methods, and when we're getting data directly from the patient and you don't have to have a site that is transcribing that data, those other methods are better today, right? And any places that we can collect the data in a digital data collection method versus EDC, the better. And I think this is really where IQVIA has put our digital-first data collection offering into place, because we do believe that someday you're going to be able to run studies without EDC. But in order to do that and to do that today, you have to figure out first where can you collect data that is a digital source and not EDC and make all of those decisions first and then say anything that's left over that I can't collect somewhere else, then maybe I can collect that in EDC or maybe I could bring in an e-source to collect that data. But that's a very, very different mindset than what people are using today because I think today, people are saying, well, I'm going to collect it in EDC and then everything else is an add-on. And I'd say we need to flip that mentality and say, what data can we collect in digital sources? And then EDC is just there for the data that can't be collected somewhere else. And if you do that today, I think you put yourself in a much better position than to actually think about removing EDC in the future. And so I know, Tony, you have a ton of experience in this and helping to guide teams. So what does that look like in a real study, and how should study teams decide where data can be captured outside of EDC? Yeah, that's a great question. So I want to lean back into one of your comments in terms of a digital-first mentality. and why go digital first? And it's a nice phrase, it's an accurate phrase, but if you look back to where it all started, and I'm very ECOA focused, and Jeff's very focused on the DHT components, but from the ECOAs standpoint, the FDA back 14 years ago, 12 years ago mandated that, they wanted to start seeing information and data collected on how the subjects are perceiving a study or how the subject are perceving their treatment and utilize that information as a part of their regulatory decision process. So right there, we started getting out of just looking at hard clinical data, lab data, lab results, x-rays, you know, PEPF meter results, all that other good stuff, and actually started getting feedback from the subjects. And how that's evolved over these years is it actually started, you know, 30 years ago, but the FDA really got interested in it. I think it was in 2011 or 2012. And so we started collecting data direct from the patients. Now, in those early days, it was all paper and transcribed it into an EDC system. That was the most common practice, but that was fraught with issues. So we have transcription errors, compliance was a nightmare. And there's lots of stories about patients who, you know, they had to do their paper diary and they're literally sitting in the parking lot of the clinic rapidly filling out their diary before they go in and see the clinician. And they were supposed to be doing it for the last month. So now we start looking at electronic ways to collect this, make it more easy, more patient-centric, where we start using digital technologies, using your phones or other types of devices. And what that really ushered in for a lot of us is more accurate data, more timely data, higher quality data. It's a little bit more complicated to collect because you got to go out to the whole patient population and be able to collect it. And how do you do that? How do you deal with the socioeconomic challenges of BYOD versus provision devices, et cetera? So we kind of have worked through a lot of those things, but the biggest benefit is the accuracy of the data, the quality of the data. And one of the things when we talk about digital first and grabbing that data, everybody starts thinking about, you know, we have the data, it's digital, we can start doing analytics on that. Now think about this way, that example I just told everybody about in terms of that patient in the parking lot, literally filling out their diaries. And what's interesting is that I've heard that story on probably more than half a dozen occasions. So it's not an uncommon, it wasn't an uncommon thing. But in a digital-first environment, we have an opportunity to take a look at that data, monitor that data in real time as it's coming in. So we know if that patient is doing their diaries as prescribed by the protocol. So as soon as they're doing their diaries, as prescribed, we know it's a higher level of accuracy of data, higher quality data when it's coming in. So focusing not only on, yes, there's more complexity, but we're getting better quality data and making sure compliance is a lot higher for the trial. So from that standpoint, transforming to digital first, it's been a transition over a decade and we're starting getting more and more momentum. We start taking a look at the digital field, at the digital health device field and all the things that we can do with, you know, PEF meters and blood glucose monitors and, you know, Bluetooth and connecting those into the other devices. I think when I listen to you talk about that and the way it started to explode and how we did kind of move from traditional methods to these digital methods, one of the challenges that it's created is how do you know when to stop with data collection. And I think, you know, we'll touch upon some of this a little later when we talk about point in time versus continuous. But Christina, right now, I mean, in the data world, it's, of course, we've got to think about how many servers do we have to hold data and things like that. But what kind of discussions are folks in trial strategies having about how much data is necessary versus how much might be too much. We can certainly collect data for data's sake, but we've got to be able to use it. Yeah, exactly. And that's one of the key pieces is that the data we collect really has to have a purpose. It has, and I come back to, you know, ensuring patient safety and making sure trial results are reliable. And so I think it's become easier and easier to collect more and more data, especially since we're not planning for that criticality and focusing really on what's important while we're building out our protocol. We don't consider our operational processes or how we're going to deliver a trial at that point in time. And I think that leads us to over-collection of data because we're sort of in a room now and we're Building EDC and we throw up a form and someone says, well, I need this field and I need that field and I needs this field. And no one's thinking back to what is it truly that we need to ensure patient safety and the results of the trial to be reliable. And so I think that is one of the key problems, as well as no one's thinking really at that point in time how much burden this is going to bring to the downstream teams that have to actually process this data. It's not just us internally who's doing data management or BIOS and reviews, but rather it's also increasing burden on how our system's set up. It's increasing burden on the patients because they're having to have more data collected, more tests done, more wearables, et cetera. It's the sites because if we're, of course, putting that data into an EDC, there's more transcription. Our study teams are impacted on the monitoring side because they have to make sure that this data is matching what's in source. And, you know, all in all, it might not ever need to be used. The biggest problem is if we collect it, we have to report on it. We have to process it. We haveto clean it. We have to report on it. So it just increases the volumes of work. And I think this contributes really to us not being able to accelerate once the study gets going. So, again, I'm going to bring it back to we have to get in there early in protocol design and start thinking about the operational elements and what data truly is required that we need to process to serve the needs of the trial and the criticality that we're trying to mitigate. Yeah, Christina, I want to jump in on one more thing there in terms of collecting too much data. And I had mentioned early on at the beginning about we need to probably start leading with some of the science and the science behind it. So some of the things that we've seen in the patient space is studies that are configured with multiple pros and multiple clinros that have duplicative concepts that they're measuring. So you're over-collecting on certain study concepts or clinical concepts that you're trying to capture. And think of it this way. It's not just a matter we have more data and then we have mehr data that we have to deal with. But now if I answer one question a certain way and I answer the exact same concept at the exact same time a different way, I now have discrepancy in my data. So it could actually cause a big problem if you have too much data in the system because the data can actually contradict each other. So now how do you solve for that particular situation? Because it's easy to collect all the data. It harkens back to exactly what you're saying. You've got to think about it at the very beginning. And not only from a system standpoint and how all these systems are put together, but from a science standpoint in terms of what you're collecting and how you're collecting it and the science behind the collection. Yeah, I think that's just so absolutely true. And I think sometimes we tend to default to doing things a certain way just because that's the way we've always done it. And we're working on a new study and we say, okay, we're just going to jump in there. We're going to collect as much as possible. and then we're going to figure out where we're going to use it all and what it's going to be used for later on like it's a question for another day and so I sometimes feel like it sounds really simple for us to say we need to put that study strategy in place up front and we needと make sure that we understand the criticality and all of the CTQs and all the RBM components and set that up front before we start to set anything up. I feel like that sounds really simple but it's actually a massive monumental shift in the way that we approach the clinical trial. And as it sits today, the way the industry is working is that we get the protocol and you are just running from day one. And we're not always giving that time to think it through and put that strategy together up front. And so I think it's all of us collectively saying we are going to do this differently. And we are going to shift our mindset because when we do, it allows us to become more proactive, more predictive, and to really be deliberate and thoughtful about the way that we are running the studies, which is going to allow us to move faster. And it is going to allow to be more efficient as well. We are constantly having pressures to figure out how to decrease the budgets of the trials. But if we are just allowing any data point to be collected, even when it's not required. We're not helping ourselves in that sense, right? And so when you really look at it, like E6R3 is pushing us down this pathway. And I hear so many people talking about the regulations and the changes and what we're being asked to do. But I hear a lot of that in terms of processes. And this is in terms that upfront data collection. I don't hear about data collection being discussed up front enough, which is what we're really pushing for here with this digital first model. And so, Christina, I kind of, you know, I know you feel passionately about this topic. And, so, you now, when you think about getting up there and being up front, but we also know that every study is different and there are timelines that are on our plates and we're, you, know, we're trying to meet all of those requirements of the study. how do you really determine what data is truly critical for a protocol and then who needs to be involved in those decisions yeah so so i think we need to like you said we need to change our approach and really do things a little bit differently even though we're under timeline constraints i think it's a conscious effort by teams to build this in it's the only way we're going to be able to accelerate, truly accelerate. And really, every study needs to be treated like a finger. It's a fingerprint. It is unique. There are critical to quality factors that are unique for every study. And E6R3 calls this out. It says to identify those factors and build that quality into both the scientific and operational design of your study. And so an R8, sorry, E6, R8. E8, R1, excuse me, says to do this as well, but without cluttering it with minor issues, unnecessary noise. And, so, really, we have to get at what's truly critical for a study. We really have to take that focused, study-specific approach and not just make it an exhaustive checklist of everything that possibly could be labeled critical. And I think that's how we approach it today. We, you know, our data strategy is, like you said, we get the protocol and we run with it. And then we have to identify everything that we need that could go wrong with the study. And we figure out how we're going to fix that or how we'RE going to, how weRE going to fix what could go wrOng instead of planning for what should go right. And to plan for what should go right and really address that criticality, you need to bring together a cross-functional team while the protocol is being developed of biostatistics, medical team members, clinical team members. Data management and digital health technology teams that are all informed by the site and the patient perspectives for that particular study and then agree on those critical to quality factors. And again, it should be like a fingerprint. It should just be what truly needs to be right for this study so that we ensure patient safety, we get a reliable interpretable results, and then define our data collection strategy at that time together to support that. And that really helps us have more focused oversight and more efficient actions during conduct as the data starts to flow in. yeah i in the i guess in the space of digital health technology when we're challenged with how much data is too much and how much collection strategy has gone too far there's two points and they're sort of good and bad digital health technologies to bring to bear on a trial are extremely expensive. So if teams are going to commit to using a particular strategy for devices, it needs to be one where it's going to impact very, very strong use of that data. It has to be a very integral part of the data acquisition strategy for typically primary and maybe secondary endpoints. And once you get into tertiary exploratory endpoints, sometimes the devices themselves can be rate-limiting because they're just too darn expensive. So that has a tendency to get people to start thinking about the use of the data and the data acquisition strategy. However, to your point, Christina, sometimes it just happens too late because then there are really strong voices, either medical voices or others, that come in to the picture downstream and really have a use case for data that can be collected in an e-source manner with digital health tech, and that could be used in decision-making for the protocol. So I guess it just underscores your point. All of the teams coming together, the sooner that we are getting together and really talking about strategy, the better. And then oftentimes tradeoffs can be made from a budgeting perspective on what's very, very necessary. But, yeah, to you, Sabrina, I know you've got a few more thoughts on this. Yeah, you know, it's and you both brought this up. So I, you know, I feel like I'm just reiterating what you've already said, which is just those different viewpoints on criticality. And the fact that it I think today, if you bring a cross functional group together, they're all going to have a different viewpoint on what is your critical data. And that's kind of part of what is creating some struggle for us. But I think it's also a source of opportunity if we think about it differently. And to me, it makes sense that teams have different viewpoints on what is critical data because they're all coming at that trial from a different perspective, right? Whether they're coming with a medical viewpoint or a biostatistics viewpoint, they are going to see the data in different ways. And so I actually think that that's a good thing. And it's something that we have to kind of embrace a little bit more but take it from happening, you know, later in the trial when you're trying to figure out what data is critical or not and shift that up front, right, and shift those discussions to the very beginning when you are putting that study strategy together. And this is really what IQVIA has done is that we have taken those experts, we have said, hey, we can help you navigate through this process so that you can truly make decisions on how to collect data in a digital first kind of method. And those experts are going to be by your side upfront, helping you to be able to make those decisions. And that's the thing that really becomes different in the way that IQB is operating differently. So that we're not just talking about this, but that we've seen results come through in a very different way than what we had before. Right. So, so I just think this part is so important. I couldn't possibly say enough, you know, that upfront study strategy is, is the place to focus. Um, but I, I know Jeff too, like one of the, the areas that IQVIA has been helping, um, sponsors for a very long time now is within connected devices. And I know your team is, ist often asked about, uh, where direct data capture or asked about direct data capture and whether it's, it's there and it can replace manual transcription and, you know, what stands in the way of broader adoption. And so just kind of wonder your thoughts on that. Yep. And I think across sponsors, across CROs, across the industry, I think there's lots of discussion going on in this space. But with respect to actually creating change to your earlier point about moving strategies forward, a lot of it started with, pulling direct data from patients with when you think about point in time collection of that endpoint data. Pretty traditional that there's a site coordinator there taking a reading from a patient and then just typing it right into EDC. And it's very simple. It's very easy. And off you go. And that's true until a bunch of different things start going on. And so, So, I mean, automation can be good. And so there has been a very tangible movement in this space where we went from point-in-time collection had to be kind of manual. It doesn't have to be anymore, whether it's vital signs or it's BMI or it' simply weight or it''s temperature or, of course, things like heart rate, et cetera. All of those point-and-time collections can be done electronically. There's no need for people to type it in and take the time to do it. Frankly, there are really advanced devices that can collect all of that stuff and simply be either electronically or, you know, they can be attached by cable and do downloads while folks are doing other things. So we prefer that everything is automated because it makes that real-time data analysis more interesting. But that critical point, when we went from point-in-time collection to much more continuous and passive data strategies, which, frankly, it's the device manufacturers that have advanced this conversation so rapidly in the last 10 years, it gives a lot more choices to what's the operational plan for your trial have to be? What is the site's responsibility? And if we're giving thought to that and we're strategizing on that, we can have some very interesting results. So it doesn't have to be extremely traditional when it comes to EDC collection. And Christine, what does that pop up for you? That just makes me think of an example that I recently heard, which didn't even pop into my mind, but it was about a six-minute walk test and how, you know, we just naturally have study coordinators with a stopwatch and having the patients do the six-minute walk test. And then they write it down on a piece of paper and then carry it over, you Know, eventually to EDC and enter it in. And that one simple test, just by wearing an activity tracker and having that information go directly to a data lake, say, that, I mean, just think of the amount of work that gets eliminated and how much more accurate that is. But it does require us to think about that method and could we do these kinds of things early on in the study because it does trickle into now we have to identify not only the devices, but we have to be able to provision them and we have get the sites trained and then we have ensure that the data flow from those devices is coming into a centralized place. So I think, you know, I find it that we just default today to EDC. I know we've said it again and again, but it's easy. Defaulting to EDCs is easy. There has to be that conscious effort required by our cross-functional team members, keeping the site and patient voice in perspective to develop that strategy. I sound like a broken record, but really it all hinges on that because it requires different conversations at the beginning of a study before we start anything. help build that into the protocol. I think, you know, part of it too is getting, when we do determine what's critical and we do determine what data we're collecting, there's going to be different kinds of risks and things we need to monitor based on the collections that we're, the collection methods we're using. So I think all of this, it really requires us to think early, early, early in the process and be ready to start that study as soon as we're ready to start setting things up rather than figuring it all out once the protocol is final and we are trying to get patients in. Yeah, I want to give one more quick example on the importance of thinking early about the study and how we manage it. We still need technology and access to technology, but not all patients are equal when it comes to technology. A great example is if you have a study where you have a geriatric population, that geriatric populace is going to handle mobile devices a whole lot differently than a 15-year-old in a pediatric study. So looking at those considerations and how not only what technology you need, but then how is the patient population going to interact with that technology could have a big impact on your ability to collect quality data. So Sabrina, let me point back to you and ask the question, what happens when all this data starts coming into a common ecosystem? Yeah, yeah. And that's, it's so interesting to think about it this way, because if you want to remove EDC from the mix, and we know that there's now, you know, on average five to 10 different digital data sources of data that are also coming in, it kind of forces you to think About the bigger picture a little bit more because people have been handling those individual data sources as transfers and they're separated. And, you know, it doesn't give you that comprehensive viewpoint against the data. And so this is where I feel like QVIA has really been at the forefront of this thinking for many, many years. We actually released our data ecosystem back in 2020. And we did so because we could see already back in like 2017, 2018, we could see this trend taking effect, right? And we could see the increase in external data that was being collected outside of EDC and to see that that data it was going to continue to increase. It wasn't, like it's a trend. It's not just a one-time thing. And so in order to bring all of the data together, you have to have a data ecosystem. um like i'd and i've been saying this now to my internal team for for well over a year is that we we're kind of moving into this new era where the question that's going to come on studies isn't what edc system are you using it's going be what data ecosystem are youusing because it's about the entirety of that data and not just edc which again is a pretty significant mind mind shift overall. But we have been bringing all of that data into our data ecosystem. It is automated. We can automate that data in, and then it gets sent to all of the downstream teams that need to be able to review it and have visualizations and dashboards against it. And what that does then is it just puts you in a really different position to beable to make decisions faster. But if we continue to allow EDC to be a large portion of that, then you're always kind of a step behind, I would say, right? You're always waiting for the site to transcribe that data, and then you have the efforts on SDV and all of the reviews because the SDV could have been incorrect. And so there's all of these extra processes that have been put into place to account for the fact that EDC data is transcribed. But all of those other digital data collection methods, that's the source data. It is what it is, and you're not going to change it. And And so when you start to make that shift and you look at the data in entirety and you don't have to put so many processes in place to ensure that the transcription was correct, that's when you get to start working in a more predictive and proactive way. Because then you're starting to look at behavior within the trial instead of just saying, oh, is this data correct? Does it match? Was it transcribed correctly? And so it's that data ecosystem that really puts us in a different position there and to be reviewing the data altogether and not just the EDC data. Yeah, and I think I wouldn't have much more to add there except to say that if we take one thing away from that, we're spending less time getting the data ready for reliable use and insights and actually being able to spend that time generating and reviewing the insights as the data is flowing in when we shift to this digital health, digital first and digital health technology data collection. So I think it just helps us transform and operate differently, as you said at the beginning of this discussion, helping us make faster decisions. In my space, when I listen to you guys talk about this, the data amalgamations, et cetera, in my space—and I think it's fair to say maybe a little bit in Tony's space, too, with our teams that are more patient data-facing—we're getting asked to bring together the subjective patient data that you might find in diaries with the objective data that your certainly collecting from digital health technologies and put that into a user experience operating system data ecosystem where people can see the operational data and make decisions on the fly. And so for us, of course, it sounds lots of folks talk about that. But to put those together in a part 11 compliant data ecosystem that, you know, my portion of the data highway might be on the front end operationally for healthcare decision makers to be able to access immediately along with diaries. But then on the back end, that part of that data ecosystem system has to be for the advanced analytics and reporting, et cetera, that's a huge undertaking. And then everybody wants it right away, like really, really, really, really fast because we have to make quicker decisions, not just on the protocol strategy side, but when there's subjects and participants on the line as well. So, Christine, I know that speed, speed, speed. When we talk about trying to reduce our dependence on traditional entities, such as EDC platforms, et cetera, a lot of that has to do with speed. Yeah, it does. And I think, you know, when we look to the future, it's a tough thing to do, to change our mindset and really think differently about how we go about setting up these trials with that strategy in mind. And so when I look to the future, I see us having, you know, three to five years out, having decision-ready data sooner, like I mentioned. We want to be able to recognize risk earlier from that data and be less reactive during our conduct processes. This is really going to help us shorten those timelines. I think what we could do today to start building this out is really, again, coming in and making the time, building in that time when we're developing protocols to really allow that cross-functional team to plan out the criticality and the data needs and how we're going to collect it and which partners we're gonna use to collect that data, because that is a discussion that needs to happen as well. And I think we wait too long to decide all of that, and it slows us down. I think EDC then becomes a deliberate choice, not a default. And that's where I think we need to start working now. I mean, but Jeff, what do you think about that? Yeah, for us, I think our responsibility here is the manufacturers of devices continue to make the form factor of devices shrink. Smaller devices, easier to use, more intuitive. And then we bring to bear oftentimes the interface between that medical grade device and the subjects or the participants in a trial, the sponsors who need access to that data, and of course, the medical professionals that are making decisions upon that data. And so we will continue to drive the intuitive use of a platform that provides user interface to expand on patient and subject diversity populations for trials, geographic proliferation for the use of trials. That means better training materials, better translation services, all of those challenging parts of bringing a more diverse patient population. Tony mentioned having pediatric populations as well as geriatric. Sometimes the automation and the use of things like mobile phones, et cetera, doesn't make it more open for different groups of subjects in trials. But you have to be able to adapt on that. And so we are continuing to drive more and more useful and diverse and rich data into the data ecosystem on a trial. Yeah, I think the science of data collection and the science actually of what's being collected is constantly evolving and it's rapidly evolving, especially as we start getting to the newer digital health technologies, other ways that we can collect data from patients. It's an evolving science. And you look back to where we were three years ago to where we are today, and you start looking forward three to five years ago from now, it's hard to predict where we're going to be. But one thing's for sure, it'll evolve. There's going to be ways to be more efficient, to be able to support the collection of data, to collect better data, higher quality data, and to support more predictive analytics for a clinical trial. Yeah, Tony, and I'm going to take it even one step further. I know it's it's hard to predict, but I'm going to do it. We're going to find out if I'm, if I're correct or not. But I believe that in three years, we will see 10 to 20% of studies that are able to run without EDC. And I think that may sound like a large amount in only three years. But I think that a lot of it is just how open and willing we are to embrace that digital data collection. And if we start to go down that path today and we startto really question ourselves at the beginning and bring in all of these experts, right, the experts that know both the pros and the cons of the different types of data collection, because there isn't any one type that's just, you know, fit for all. And you're like, oh, this is a slam dunk. Go make it happen. That doesn't exist. You have to be able to understand that protocol and make the right decisions for the protocol. And with the right experts there to help guide you through that, it is possible. But we have to start today because I don't think that anybody is comfortable flipping the light switch here. I don' t think we' re going to see people just jump in there and be like, okay, I got it, like run a study without EDC. I think it' s going to have to do a progression where the steps are small, but then when you look at the big picture, it shows you just how far you've come in three years. And if we're willing to go down that path, I think it is absolutely possible to see 10 to 20% of studies without EDC in three ears. All right. I think that brings us to the end of our panel today. And I hope that this was beneficial to everybody and that you got a better sense of how IQV had really seen data collection opportunities to do things differently as we continue forward. And so I think now I'm going to turn it back to Gina because we're going to answer some questions. Thank you. You know, just from my standpoint, fantastic conversation over the last hour. And I have to tell you, there were so many times I wanted to interject and be a part of the discussion. But, you know, there were a couple of takeaways from today's discussion that I thought were so important just to bring up and remind everyone real quick. First, everyone, this whole field is just so complex because even from what you each addressed from your own perspective today, I feel like everybody is looking for something a little bit different. Everyone has very unique needs, which complicate things. They make things very interesting. Tony, something that you brought up right at the end with your considerations about perspectives moving in the future, you said, you know, in the last three years, we've evolved. It's funny you say this, because just this morning, I was working on something. And I was thinking about how far we've come just this year alone. You know, everything is evolving really quick. And things that we thought weren't possible just six months ago, if we would have had this conversation back in January of 2026, and then repeat this now where things have changed and then you know sabrina it makes me almost have the challenge will it take three years to get to that that percentage that you're saying i see christina you know smiling too like oh you know challenge accepted but all of this leads me to just saying um that's why i really truly appreciate these roundtable type discussions because i just want to bring these conversations back again and have you join us again to to explore where we are where we are this time next year. But that said, in interest of everybody's time, yes, let's jump right over to the Q&A session today. Again, to our audience, thank you for your participation. A lot of questions in the queue, some really great ones that I want to make sure we address. But, that said don't feel discouraged if we don't have a chance to respond live. We are almost out of time. Please continue to submit any questions or even perspective that you may have right into the Q&A chat, and our subject matter experts will be fielding and following up following today's broadcast. All right, time for the hot seat. A couple minutes left. With the move to a digital-first approach where all data is collected directly and EDC is an add-on, how do you all foresee the management of that data? Is it a data lake offering, an upgraded platform? Is it going to be like EDC 2.0? What do you think? Yeah, I can take that one. So I think it is. I think to get to full EDC list, it is pushing some of those data ecosystem concepts that are out there. And IQV has them. There's commercially available data ecosystems as well, whether they're based on data lake or other technology. And I think that the data ecosystems that are available, I think they are absolutely critical to having an EDC-less trial strategy or a digital first data collection strategy. And in order to get us to push down that path more and to be able to run studies without EDC, I do think that we have to continue to upgrade those, give more functionality so that everybody can see all of that data together in one place. I think we have a lot of really great visualizations that bring the data together, but I think there's going to be more requests for different ways to be able to view that data all in one central place. And so I do think that there is more needed from a technology standpoint, but i also think that there's great opportunity for us to move to digital first today while that technology is evolving. Great. Thank you, Sabrina. All right. and jeannie i wonder if we should skip down uh to the to i see several uh iqvia questions which are really intriguing as well um but then a couple of external um guests have asked questions i'm i don't know if anybody on the panel has some thoughts but i'm happy to take a crack at at pk and herman's um questions there and maybe you guys can chime in but i love uh pk's question about. So with all this great automation talk, how come the FDA isn't releasing more drugs more quickly? That's a challenging question. Probably way before my pay grade, but I will say this, that I have oftentimes challenged our team with the same sort of questions. And And again, I don't want to quote data here, but maybe, PK, what we do sometimes as ourselves and the professionals of this business is look at the horizon a bit too short of a manner. And I think if we go back maybe 10 years, I think the FDA actually is approving quite a few more drugs than they did a little bit longer. So maybe we are making an impact. But I also believe, and I'm sure everybody agrees, that there is no arguing with the level of automation that new tools have given us now and the fierce pace with which automation and, you know, I think AI is a bit of an overused term these days, but nonetheless, artificial intelligence and other types of automation is absolutely, we are at breakneck speed now. So I would venture a guess that we'll continue to see more and more drugs released more rapidly. I would also add that from our own experiences with our own sponsor customers, not always on a drug release, but certainly on additional study strategies where we're running more trials and different types of trials as a result of a previous trial. We're certainly seeing a lot more of that. So trial activity, I think, is going up as well. So just a couple of thoughts there. I don't know if anybody else on the panel has an additional add. Yeah, I agree with everything that she said there, Jeff. And I'd say that the technology is definitely helping us, but it feels like we haven't taken the next step that comes after the introduction of technology, which is to look at what is the net effect of all of that different technology and our processes and the way that we work. And I think until we get there and we're really willing to adjust processes and look at roles that can be, you know, put together in different ways, I think we will struggle with this some. But the amount of data that we collect is directly attributable to how much work we have to do on the trials. So that's something that I think we can really start to get after today. And sorry, I think I talked over you. Sorry, Sabrina. No, you said it. You covered what I was going to say, but there's one piece I can add there is I think that the increase of the data, I mean, we have all this technology and automation and so forth, but it doesn't change the fact that when we do use one of these digital first sources on a study and still have EDC, there's a tendency to still collect information about that digital source in EDC. So we're still, we have that source, plus we have transcription, you know, transcription of some information and reconciliation and monitoring and cleaning and analysis and all of that stuff has to happen. And so I think that, you know really it comes down to figuring out what's right for the study and then relying on those digital health or digital first collection methods to give us the data we need without having to capture extraneous information or duplicative information elsewhere. I think that's a perfect way to wrap up today. I think we literally just addressed about four questions in the queue with that wrap up, but, you know, all important questions. I'm going to leave some of them in the cue for you guys to follow up on after, because I know there were some additional points and perspectives in there that I'm sure that you'd like to follow up on. We are going to wrap up. This was a really great use of my time personally, and I hope that our audience feels that way. I want to thank our entire panel for making today's time valuable for us and for participating. I know it's not easy on everybody's schedule to put this together and to produce this. So I want thank you for all of the great insights today. I want to thank the great questions that came through the participation from our audience. And I'd like to ask if our audience wouldn't mind participating in a brief survey. The survey just helps us try to collect some information, some insights on how to design future programs that are worth your time and of your interest, such as today's. So we hope you'll take a moment to fill that out. You're also going to receive an email alerting you once today's discussion is available on demand for replay. We hope that you'll kindly pass that along to any colleagues, friends that you think would find this conversation interesting and valuable, too, as well. With that, I'm going to wrap up. I'm gonna give everybody the rest of their day back. But thank you so much, everybody, for participating today, and we hope that you'll join us again soon. Again, I kind of hope we have a follow-up discussion in the next year where we discuss a similar topic once again and see where we're at. And Sabrina, we're gonna hold you to that three-year goal. Yes, I'd love to come back and we can assess all the ways that we beat that prediction. Again, thank you to everybody. Have a wonderful rest of your day. Take care. Thank you. Thank you, Dr. Kahn.