X

 

 

The Quest to Eliminate the Data Silos In Life Sciences
Episode

Arnaub Chatterjee, Senior Vice President of Product and Ecosystem at Acorn AI at Medidata Solutions

The Quest to Eliminate the Data Silos In Life Sciences

In this episode, we have the privilege of hosting the outstanding Arnaub Chatterjee, Senior Vice President of Product and Ecosystem at Acorn AI at Medidata Solutions. Arnaub discusses how his company manages and builds products off of historical clinical trial data to assist the biopharma space. He educates us into the complicated pharma data world, and how the silos between clinical data and real-world data are making it difficult to see and understand the whole patient experience which is crucial in building a better drug. Arnaub also shares some successes, including using a synthetic patient population for phase 3 studies without having actual patients go through a very invasive trial. There are so many things to learn from this amazing interview on clinical trials, AI, and patient data, so please tune in!

The Quest to Eliminate the Data Silos In Life Sciences

About Arnaub Chatterjee

Arnaub is the Senior Vice President of product and ecosystem at Acorn AI at Medidata Solutions Company. Medidata is a global provider of cloud-based and analytic solutions and life sciences. And in his role, he leads the development and go-to market of data science products within the company. In addition, in his to his role at Medidata, he also serves as a Teaching Associate in the Department of Health Care Policy at Harvard Medical School and lecturer in the Department of Policy Analysis and Management at Cornell University. Prior to Medidata, Arnaub was an associate partner in the Pharmaceutical and Medical Products Group at McKinsey and Company, where he advised pharmaceutical and technology companies on a range of topics, including entering new markets, utilizing novel data and analytics, and digital transformation. Before his time at McKinsey, he served as director of Merck’s Data Science and Insights Group, where he led ventures and partnerships for Merck’s Outcomes Research Group, Topower Research and Development and commercial activities. He previously served in the Obama administration as an Adviser to Former Chief Technology officers Todd Park and Brian Savic at the US Department of Health Care and Human Services. And he also worked in the Office of the Secretary at HHS as a Lead Policy Analyst on health care fraud and abuse initiatives around the Affordable Care Act. 

The Quest to Eliminate the Data Silos In Life Sciences with Arnaub Chatterjee, Senior Vice President of Product and Ecosystem at Acorn AI at Medidata Solutions: Audio automatically transcribed by Sonix

The Quest to Eliminate the Data Silos In Life Sciences with Arnaub Chatterjee, Senior Vice President of Product and Ecosystem at Acorn AI at Medidata Solutions: this mp3 audio file was automatically transcribed by Sonix with the best speech-to-text algorithms. This transcript may contain errors.

Saul Marquez:
Hey Outcomes Rocket listeners, Saul Marquez here. I get what a phenomenal asset a podcast could be for your business and also how frustrating it is to navigate editing and production, monetization, and achieving the ROI you’re looking for. Technical busywork shouldn’t stop you from getting your genius into the world, though. You should be able to build your brand easily with a professional podcast that gets attention. A patched-up podcast could ruin your business. Let us do the technical busy work behind the scenes while you share your genius on the mic and take the industry stage. Visit smoothpodcasting.com to learn more. That’s smoothpodcasting.com to learn more.

Saul Marquez:
Welcome back to Outcomes Rocket everyone, Saul Marquez here. Today, I have the privilege of hosting the outstanding Arnaub Chatterjee. He’s the Senior Vice President of product and ecosystem at Acorn AI at Medidata Solutions Company. Medidata is a global provider of cloud-based and analytic solutions and life sciences. And in his role, he leads the development and go-to market of data science products within the company. In addition, in his to his role at Medidata, he also serves as Teaching Associate in the Department of Health Care Policy at Harvard Medical School and lecturer in the Department of Policy Analysis and Management at Cornell University. Prior to Medidata, Arnaub was an associate partner in the Pharmaceutical and Medical Products Group at McKinsey and Company, where he advised pharmaceutical and technology companies on a range of topics, including entering new markets, utilizing novel data and analytics, and digital transformation. Before his time at McKinsey, he served as director of Merck’s Data Science and Insights Group, where he led ventures and partnerships for Merck’s Outcomes Research Group, Topower Research and Development and commercial activities. He previously served in the Obama administration as an Adviser to Former Chief Technology officers Todd Park and Brian Savic at the US Department of Health Care and Human Services. And he also worked in the Office of the Secretary at HHS as a Lead Policy Analyst on health care fraud and abuse initiatives around the Affordable Care Act.

Saul Marquez:
Just an extraordinary contributor to our health care space and a forward-thinking innovator, really privileged to have him on the podcast today. Arnaub, thanks so much for joining us.

Arnaub Chatterjee:
Yeah, thanks for the opportunity. Great to be with you guys.

Saul Marquez:
Yeah. So talk to us a little bit about you Arnaub. What is it about health care that inspires you to stay focused in the field?

Arnaub Chatterjee:
Sure. So I guess if I start up on a personal note, I can say that medicine and health care are very much embedded in my DNA. I come from a line of physicians that spans multiple generations and grew up with the stories of different patient encounters with different clinical settings. So everyone my grandfather, my father, my sister, my brother-in-law are all either physicians or health services researchers or both. So you could imagine that our Thanksgiving discussions are pretty much-heated conversations over the state of health policy today. You know, aside from my family, I’ve had the opportunity to see across the health care system in various roles over the last, I guess, 10, 12 years now. And I’m kind of had a bite and consulting and pharma in the government space and in academia. And I think the thing that keeps me going is that I’ve been fortunate to be part of what I call the health care movement. And these are kind of pivotal changes or sort of tectonic shifts in our health care system that have happened within the last decade and kind of fundamentally transform the industry. But also kind of my thinking about how the health care system as a whole is evolving. So some of the stuff that you mentioned in my bio, whether it was working on the Affordable Care Act, which was such an important piece of legislation, or being part of some of these larger data and technology movements, even through the lens of big government, big things that happened in over the last several years.

Arnaub Chatterjee:
And then more recently when I was at Merck, I had a chance to better understand what’s kind of commonly called now as real-world data, which is everything happening outside of data and clinical trials. And could that tie into improving economics or outcomes research within that company? And I guess my most recent slate of experiences are really pushing towards how do you kind of move the needle in pharma research and development and how do you better understand where data science and technology intersect with that changing space? So kind of the totality of everything. If you think about how interconnected the system is, having those experiences, I think have shaped my thinking now and really to where we are today. So that’s what it’s been, fun, intense, and kind of an inspirational experience today for me. And I’m excited to continue to develop and learn.

Saul Marquez:
That’s fantastic, Arnaub and it’s at your core. And the experience that you’ve built is certainly diverse in its base and from commercial markets to dealing directly with pharma and even in health policy. Just fascinating work that you’re doing. And so today, your work, Acorn AI is really focused on delivering actionable insights, right? So talk to us a little bit about what that means and how you guys are adding value to the health care ecosystem.

Arnaub Chatterjee:
Sure. To provide a bit of background, Medidata is a 20-year-old electronic data capture company that really came of age during the SAS and cloud technology eras. And at that time in the 90s, when the co-founders targeted Glenn were conceiving this company, they were thinking that this notion of capturing data from the clinical trial is going to be really important and their bet was right. And as a platform, Medidata, now in our 20th year, captures about and manages more than 50 percent of the world’s clinical trial data that are on these digital platforms. At the company we have different products now that sit across sort of the entire spectrum of clinical trials and from data management all the way through the advanced analytics. So about a year and a half ago, Medidata was acquired by Dassault Systèmes, which is a Parisian-based company. It’s one of the world’s leading companies in modeling and simulation. This kind of puts us in a unique position to help think about how do you build this future of evidence and evidence generation that really assists not only the biopharma space but also med device, diagnostic companies and even academia as they think about what value is for drug development. How do you minimize risk and what is outcomes actually mean? So to your point, ACORN is Medidata’s data science entity. And I think what makes us interesting and unique in the space is that the data that we’re managing and building our products off of is actually historical clinical trial data that we are managing on behalf of our sponsors that we work with.

Arnaub Chatterjee:
We also have privileged access and data rights to that historical data. So this is kind of the largest data set of its kind right now in the world. It’s about six point three million patients. It’s twenty-two thousand trials worth of clinical trial data. And our vision is to take this trial data and better understand what happened so that we can improve drug development.

Arnaub Chatterjee:
And can we take what’s happening within trial and better understand how patients are actually responding to therapies or treatments in what’s called the real world, what happens after a clinical trial. So that’s how we think about our spot in the world and kind of what our focus and our vision is.

Saul Marquez:
Well, with a data set and and history, like you mentioned, I mean, 22,000 trials. I mean, how many patients did you say? 63 million?

Arnaub Chatterjee:
Six point three million patients.

Saul Marquez:
6.3 million patients. Just I mean, wow. And when you think about building models, you need this richness of data. Certainly exciting to hear the play that you guys are after. What would you say makes what you guys do and offer unique in the market?

Arnaub Chatterjee:
Yeah, I think that maybe I can provide a bit of context for the audience into if you allow me to just for a second here about how messy and complicated the pharma data world is. So right now you have these kinds of two main, I guess, buckets of data. And one of them is clinical trial data, which is by and large, what a pharma company might use internally. It’s their own trial data that they’ve been managing for years, or they get this information from published literature. So they have some benchmark of a published trial to get a better sense of how to design their own trial. So that’s kind of one bucket. And the other bucket is what I was mentioning earlier. This is real world, which is everything outside of that clinical trial. And that could be a billing records and insurance company. It could be a medical record from a doctor or it could be even a genomic record from a DNA. And I think the Holy Grail here is to better understand how a patient did in and out of the trial. And that’s how you really understand if the drug is safe and effective. But right now, the problem that we’re trying to address and so forth is that these two buckets live in completely different silos, as you can imagine. And the data is really hard to connect between the trial and what happens to that patient after the trial. So to your question around what makes us different than what’s available today, maybe I can present kind of like two different visions here where, you know, if we look at all of this historical information and we’re able to capture all of the endpoints and all of the covariates that were ever measured in previous trials, and that includes something like a biomarker.

Arnaub Chatterjee:
It includes all the outcomes. And if they all appeared on our trial in parallel, we can also see everything that happened operationally. So we know what’s happening with a site or we know what’s happening with a P.I. and investigator. And we have an almost Real-Time view of performance because clinical trials also have an operational component. That totality of evidence in our minds gives us a better confidence and validation in the way that biopharma or device or other companies design their trial. And if you can create a sharper benchmark, whether it’s on recruitment or whether it’s around designing your cloud more effectively, that we think is a game-changer with this data set. So that’s sort of the first half of the vision. The second half is that if we have the trial data and we’re able to better understand what’s happening in the so-called real world, the regulatory world, like the FDA, even biopharma themselves, like want to have a better understanding of what to make of this data and how to interpret it. So right now, sort of our future state is to build pipes between these patients so it can be solved who’s within a clinical trial. And if we can connect Saul’s data from his trial to his billing record or to his medical record, that means we have the entire patient experience. If you can do this, you’re really redesigning the way you think about trials and the way your medicines may or may not be effective or safe, but you’re kind of trying to reprogram that entire process from start to finish.

Saul Marquez:
Yeah, that’s really, really interesting. And the fact that these two data sets, the clinical trial data and real world data, as you’re calling it today, I mean, they should be integrated. And why has it not in the past or not well past?

Arnaub Chatterjee:
Because clinical trials are a very focused set of data that’s being captured, as you can imagine. Trials have very specific criteria for measuring a very specific endpoint or an outcome. And for that reason, that’s how you understand whether a drug is truly safe or efficacious. So what’s called a randomized control trial has been the gold standard for how drugs are developed for as long as people can remember. But I’d say over the last decade or so, this growth of new data has really changed the way that we interpret what good looks like.

Arnaub Chatterjee:
And I think those data just exist everywhere from sensors. And what’s happening on your phone to how are you kind of interacting with an insurance company? All of that is valuable medical information as well and adds a lot of context towards behaviors and patient perceptions and more of the different kinds of qualitative and quantitative measures of performance and health. And I think that is in a whole different bucket. So our thought is that this is coming a time now where we need to be able to see across the entirety of that experience to really understand if we can build a better drug.

Saul Marquez:
Yeah, and just thinking through it’s about optimization. It’s the clinical trial is about getting from point A to point B with a therapy in a health state. But you’re talking about how do we get there more efficiently, less costly, more convenient to the patient and so on. That train of thought, how would you say what you guys are doing is improving outcomes or even making business better?

Arnaub Chatterjee:
Sure. So let me share maybe a few different examples here. And if you can remember from what I just said, like our vision is sort of broken down into two pieces. The first part of the vision is around how do we repurpose and reuse our historical trial data? The second purpose is around what do we do when we collect data together? So for that first example, we have a product called Synthetic Control Arms. And the idea is that you can reuse or repurpose historical data that can augment or even replace the control arm of a trial. And what that does is that it reduces the need to recruit real prospective patients into the control arm. And this is important for indications or diseases where patients view the standard of care control therapy as not desirable. And what we’re saying is that that process of finding patients for that control arm is timely and it’s consuming. So what we did recently was we used our historical trial data and we actually were able to make that as part of the actual submission for a clinical trial, meaning that the companies that we worked with didn’t have to recruit certain patients because they were able to use historical data. So you sort of have this synthetic patient population that was used. And we did this for a disease that pretty tragically did it for phase three studies for recurrent glioblastoma. And this is a really debilitating disease for patients.

Arnaub Chatterjee:
And it has a prognosis of less than six months. Many patients die during the trial. And the other thing is that the treatments for these patients are very, very invasive. It typically involves injecting a therapy into the brain tumor by drilling a hole directly into the skull. So if you can imagine the value of taking historical patients who’ve already been part of a previous trial, they look very similar to these patients by their data for this new trial that you’re trying to recruit for. If you don’t have to get those patients recruited in that trial, you’re saving anywhere four to six months of time for this disease where many patients actually die during that trial. So for us to get not only one but two regulatory successes on the design of these trials, using our data and using this method is a pretty big move for the industry and something that we were fortunate that the regulatory bodies like the FDA were supportive of. So that’s one example. The other one real quick, is around work that we’re doing in COVID. And this is going to the second part of our vision. This is around linking data. So we know that COVID vaccines are kind of a miracle of science because vaccines take a long time to develop. You need a lot of safety and efficacy data. And even though we have these vaccines now, covid is still kind of just it’s just months old, really.

Arnaub Chatterjee:
It’s something that has a hodgepodge of symptoms and severities and there’s still a lot of unknowns. So we really, in an ideal world, would want one to three years of follow-up to really capture adequate safety and efficacy. And our current capture, the trial data is giving us some glimpse into whether the vaccines are safe or efficacious. But it’s still early days. And as I was mentioning earlier, we don’t really know how patients are going to do in the real world. We don’t know what they look like. We don’t have that long period of safety or efficacy data to understand these vaccines in their. So what we’re doing actually is we’re working on with biopharma sponsors right now and we’re actually linking data from an ongoing covid trial for some of the biggest pharma companies out there to patients, medical records and other data. So that means that if you knew about a patient’s comorbidities or their previous interactions with the health care system, you could better understand whether a safety event is real or not. And you’re kind of prospectively designing this study to better understand how patients are going to put. Armed with a vaccine. So right now, over a period of time, we’ll have captured a large population of linked patients will be able to stratify their outcomes and then we’ll have better certainty on what’s working and what’s not working.

Saul Marquez:
Well, two phenomenal applications and just fantastic and congratulations that you guys were able to do two successful submissions using historical data for that control arm. I mean, just like I was taking a little note and I wrote Boom with a big exclamation mark because it’s, you know, like I’m just like this is yeah.

Arnaub Chatterjee:
That’s how we feel.

Saul Marquez:
To be able to do that is just and then the severity Right. and of the patient condition and the invasiveness of the approach and the and the saved just like just phenomenal. So so first of all, congrats on that. And I mean, huge kudos to our FDA group too, for actually seeing that and being able to say, yeah, this makes sense.

Arnaub Chatterjee:
Thank you for that. And absolutely, you have to work hand-in-hand with regulatory bodies to help them understand where applications like this synthetic population makes sense. And it is for these debilitating diseases. It is for the ones that have challenging recruitment timelines. So it’s not a silver bullet for everything. But there are certainly areas where if you have real unmet medical need, you can apply stuff like this.

Saul Marquez:
Wow. So cool. Very, very cool. I’m sure everybody listening to this, you’re like, wow. Yeah, for sure. This is amazing and a testament to what’s possible, you know, and so the beginning of something great. So as you think about the challenges, holy smokes. I mean, I can’t imagine the stuff that you guys have had to deal with. So talk to us about maybe one of those setbacks and a key learning that came out of it.

Arnaub Chatterjee:
Yeah, I might get existential for a second here and actually think about a larger issue as it relates to some of these novel techniques and ways of using data. And maybe that from the core thesis to start with here is that the drug development problem still exists and it still persists. Right. We’re still spending more money and time to develop the drug and the cost of these failures are increasing. So on one side, you have data liquidity that’s gone up and you have incredible processing capabilities that have increased over time. But we still come back to the fundamental question of like, how have we moved the needle? So if you said my shoes or other technology company shoes, the key question comes down to how you demonstrate value in data science and technology.

Arnaub Chatterjee:
And if you think about the journey of AI and machine learning and how that’s impacted drug development, I talked to a number of colleagues on a given day on what needles to be moved and how. So what you’re trying to do and if you’re in a tech company shoes is kind of explain what is the demonstrable correlation between something you do like a data science method or activity and how that meaningfully affected cost or time or some other tangible metric. And I think some of the real learnings that have come out of that have been interesting. The fact is that like the biggest pharma goes out there have made really big bets on data science, but there’s still a really established process in drug development. And their stage is that you have to hit, like you mentioned, you’ve got to go from a hit to a lead, a lead to a development, development to commercialization. And what data science sometimes does is it challenges these assumptions and these hypotheses. And sometimes that flies against the face of really deeply rooted preconceptions that are held in these companies. And the real proof points that the onus is on a tech company to show that you push the thinking, you shorten the timeline, you kind of alter the course or in action. And I think that’s that. There’s still a healthy dose of skepticism in the big world of biopharma that that data science can or cannot do those things. The second thing I’ll say is there’s a funny quote from Ventor pioneer Marc Andreessen, who once famously said that software is going to get the world.

Arnaub Chatterjee:
And then he most recently said that if software eats the world and biology actually eats software, and what that means is like biology is incredibly complex. We have this great opportunity in front of us to better understand what to make of all this new data coming out. But there is a world now where data science has never been more important to process all this information. But we also live in this highly complex world of biology and we have to operate under this garbage in, garbage out mentality for how we apply data science. So, you know, the other big learning here is that your models for doing this are only as good as what goes into it. And your data has to be epidemiologically representative and clinically valid in order for it to be a true method that people can get behind. So that’s just the other things, is that’s what we have a lot of faith in the data that we’re using, because it’s sort of the basis for all these other trials. But I think overall, if you take all that into account, the stuff that’s cool is that there are all these new biotechs that are starting up, that are approaching drug development with a data-first mentality and a data science first mentality.

Arnaub Chatterjee:
And these are companies built from the ground up with data science where they’re taking all of that biology domain expertise and they’re applying it with data science expertise and that changes their entire development pipeline. So a lot to look forward to, I think, even despite some of those setbacks I mentioned.

Saul Marquez:
Yeah, I totally agree. And yeah, you call that a really, really challenging. Is that the onus ends up on the tech company and in a very cynical, deeply embedded process? Nevertheless, I mean, examples like the one you just shared, I think you really it’s clear Right. and I don’t want to say you can’t argue with it, but if you say four to six months of time and you’re able to use historical data for a submission for that control arm, I think that there’s a demonstrable correlation. That’s my perspective.

Arnaub Chatterjee:
I think you’re right. I think it’s you got to land. You’ve got to land these whales right. And you’ve got to find these types of examples. And those aren’t easy to come by. You kind of go through a bunch of different projects and exercises and prove points until you can find examples like that. So you’re totally right.

Saul Marquez:
And so what would you say you’re most excited about today?

Arnaub Chatterjee:
So I think the idea of the ability to link data, it sounds like an operational problem and in some ways it is. But we have to remember that the core value about what it means for the patient is pretty phenomenal. So the data at its core, whether it’s a clinical trial or medical record, is the patient. And we have to remember that everything we’re trying to do is to improve that patient outcome, that patient experience. So why I’m excited about some of our data science products are kind of what you mentioned. It’s using these synthetic patient populations and they don’t always have to be homerun use cases like shortening a trial timeline or replacing a control arm entirely. There’s a lot you can do that incrementally changes the way you can power a smarter trial in a more expeditious trial. And that’s been a goal of the industry for a long time. If you have a few regulatory successes under your belt, like we have, I think we can show that we can change those timelines. But also, can you meaningfully alter the course of clinical development in a way where you’re embedded in workflows and you’re kind of changing the way that these patients are ultimately getting access to different and better treatments? And I think that’s kind of an important part of the future. I think the other thing is that for the patient, that if you have this longitudinal record where you can kind of document the entirety of their experience, that data feedback loop could give you the most comprehensive look into patient responsiveness to treatment.

Arnaub Chatterjee:
And that should come back to drug development. It should inform the smarter treatment. It should design the trial differently. So I think for biopharma, the upside is massive if you get this right, which is the ability to change and enhance your trial design by finally being able to leverage some of that clinical trial data. So that’s kind of one or two things that excited about. The other stuff is on the regulatory side, going back to the FDA and others for a second, I think this means like a pretty big advancement in the way we think about long-term monitoring and safety surveillance. And this has immediate significance during COVID, because if we can start to show these COVID trials, which were very short in nature, the concept of linking patients together and understanding those longitudinal insights could change the way we think about drug safety for some time. So I’m hopeful that regulatory bodies are responsive to this new methodology and this new way of data collection. And ultimately, you know, I think it’ll help us understand the context of what’s unique to the trial and what’s unique to the real world and how to use those data sets together. So for me, if we’re the company that’s got to position that the initiation of so many clinical trials and we kind of manage the data for all of these trials, creating this feedback loop is very much within grasp of that.

Saul Marquez:
Yeah, it’s certainly exciting. Something to look forward to. Really appreciate you sharing all of the amazing work that you and the team at ACORN are up to. Folks, it’s Medidate.com/AcornAI. Or just go to our website OutcomesRocket.health type in AcornAI, you’ll find the full transcript and show notes of our discussion with our Not today. Now, why don’t you leave us with a closing thought and what you believe is the best place for people to reach out to you or learn more about what you guys have to offer?

Arnaub Chatterjee:
Yeah, great question. You know, as we’ve kind of developed and thought about the growth of our business and where we sit within the ecosystem of biopharma, I guess one recent observation I just had myself is probably just as a company, is never to kind of remain static and constantly evolve. And what that means is really understand, like how your client or how you’re patient or how your stakeholders’ needs are changing and evolving and get firsthand feedback and better understand the problems that they’re trying to solve. Then there’s really nothing quite like adaptation as a business. And then one day hitting a stride with a problem like a synthetic control arm that you’re able to solve consistently and prove that you have differentiation. And in my humble opinion, the best way to do that, I think, is to continue to understand how whatever it is that you’re building kind of fits into this pre-existing world that’s full of people and personalities and products and data knowledge. And you kind of have to show that you’re able to make some of those things better or kind of enhance the way that people are using existing solutions. So I’m more than happy to chat with folks. They can reach out to me on LinkedIn and it would be welcome the opportunity to connect and be outstanding.

Saul Marquez:
And again, folks, just. Go to the website OutcomesRocket.Health type in AcornAI, you’ll find all the links, including a link to Arnaub’ss LinkedIn profile so you could connect, then start a conversation there. Something today resonated with you. I know it definitely resonated with me, so can’t thank you enough or not for the work you and your team do and for spending time with us here on the OutcomesRocket. Really appreciate you.

Arnaub Chatterjee:
Yeah. Thanks so much for the update, as you can guess. Enjoyed it.

Saul Marquez:
Hey everyone, Saul Marquez here. Have you launched your podcast already and discovered what a pain it can be to keep up with editing, production, show notes, transcripts and operations? What if you could turn over the keys to your podcast busywork while you do the fun stuff like expanding your network and taking the industry stage? Let us edit your first episode for free so you can experience the freedom. Visit smoothpodcasting.com to learn more. That’s smoothpodcasting.com to learn more.

Sonix is the world’s most advanced automated transcription, translation, and subtitling platform. Fast, accurate, and affordable.

Automatically convert your mp3 files to text (txt file), Microsoft Word (docx file), and SubRip Subtitle (srt file) in minutes.

Sonix has many features that you’d love including powerful integrations and APIs, collaboration tools, transcribe multiple languages, automated transcription, and easily transcribe your Zoom meetings. Try Sonix for free today.


Things You’ll Learn

  • You have to work hand-in-hand with regulatory bodies to help them understand where applications like this synthetic population make sense.
  • Trials have very specific criteria for measuring a very specific endpoint or an outcome.
  • Randomized control trial has been the gold standard for how drugs are developed for as long as people can remember.
  • What data science sometimes does is it challenges these assumptions and these hypotheses. And sometimes that flies against the face of really deeply rooted preconceptions.
  • Your data has to be epidemiologically representative and clinically valid in order for it to be a true method that people can get behind.
  • Remember that everything we’re trying to do is to improve that patient outcome, that patient experience. 

 

Resources

https://www.medidata.com/en/

https://www.medidata.com/en/clinical-trial-products/acorn-ai

https://www.linkedin.com/in/arnaub/

Visit US HERE