UCI Podcast: Putting data to work through collaboratories

(00:09)

From the university of California, Irvine I’m Sheri Ledbetter and you’re listening to the UCI Podcast. Today I’m speaking with Tom Andriola, vice chancellor for information technology and data and chief data officer at UCI. Today, we’re going to discuss the role of data analytics and data science in the advancement of health science in the form of an initiative known as a collaboratory. Tom, thank you for joining me today,

(00:36)

Sheri, thank you for having me.

(00:38)

Let’s start with the basics. What is a collaboratory?

(00:42)

So you can find a formal definition out there for a collaboratory. I like to talk about it in terms of a boundary-less organizational entity that creates rich recurring interactions between subject matter experts, where the commonality is they all want to work with data and do interesting things with it. And that might be research, that might be doing innovation, it might be working with industry at UCI. We’ve  honed down that definition a little more specifically, and we say the collaboratory is about putting our data to work, to create value for our different stakeholders. That means internally, if you look at great organizations internally, they use data to consistently make good decisions and do so at increasing value. That’s like the way Amazon looks at it, but then also we have this opportunity to take that same data and use it externally.

(01:36)

It becomes mechanisms for our researchers to be at the top of their game, by taking our internal data assets, de-identifying them, and then making it a research asset that they propose to a grant to NSF or NIH. It’s something that we take and actually work with other institutions and combine their data and our data, and actually do research together or push the boundary of a field. And we’ll talk a little bit about how we’re using it in the healthcare setting to create really transformational changes to the way that we care for patients or help our students succeed.

(02:09)

That’s interesting. I like how you talk about putting data to work, how our collaboratory is disrupting the traditional data analytics model.

(02:17)

I would call it not disrupting as much as evolving. I think our understanding of data as an asset is evolving to be more than just an asset, you know. You may have heard the term data is the new oil. And you know, I really don’t like that paradigm because I actually don’t think it fits. I mean, I get the concept behind it, but I’ve actually subscribed to, ‘data is like water.’ First of all, it’s everywhere. Now think about all the things that are now captured. Also, you also do digitally, you don’t look at a paper map, you look at an app and it helps you get to where you want to go. It also is tracking how long you’re out. So it actually then uses that data to go back and use for how long it takes to get from here to there for the next person who does the search, right?

(03:00)

So data is everywhere. And the reason I like water as the analogy is you have to capture the data. You have to channel it. Data can be used for multiple purposes. If I capture the rainwater coming off of my roof in a gutter and into a drain, I can use it to water the plants. I can use it to feed the animals. I can use it for basically lots of different things, right? So data is everywhere. The whole key in today’s world is to think of it, not just as a one purpose thing, but that can be used for multiple purposes. So I like to say data is an asset. Yes, but it’s also a currency because you can exchange it and work with people with it. It’s an ecosystem. You can use it to pull people together. And overall, it’s actually now a strategy. You have to think of data as a strategy for your organization and what they do. And that’s how we’re thinking about it here.

(03:52)

That’s really interesting to think of it as a currency, as a strategy. So many different things.

(03:59)

Let me add a little bit more. It’s, you know, data needs to be free. It’s the way we think about it. Like data needs to be free and it needs to be fluid means it needs to be able to move in exchange without friction. We have to think in terms of platform, so bringing different types of data together, you know, if you talk to Google, it’s all about more data, more data. What we like is diverse data, right? In the healthcare context, it’s electronic health records and medical images and your genomic profile and your EKG wave forms – all different types of data together. So it has to be multimodal and ultimately it has to be curated or contextualized. That’s when it’s the most value is, you know, data just thrown into a blob isn’t very valuable. If you curate it and give it definition and meaning it gets more valuable. If you contextualize it around what you want to do with it, then that’s when it’s the most valuable. It’s like the difference between just the rainwater and the water that goes through your filter that you ultimately drink, it’s contextualized for a single purpose. That’s when the data can come up really, really powerful for us.

(05:01)

Yeah, powerful is what I was thinking. How do you see the impact of the collaboratories long-term on medical or data or patient care advancements then?

(05:11)

So here at UCI, under the College of Health Sciences and Health Affairs, we have this term, we say discover, teach heal. And so I think what you’re going to see is data play a role in really advancing all three of those, right? So discovery is the research side, our data assets and how we work with other organizations to combine their data assets with our data assets is pushing the boundaries of how we  think about doing research of disease, research of patient journeys, research into the effectiveness of different types of drugs and therapies, teaches about our mission of we educate the workforce of tomorrow. And so in a healthcare context, what that means is how do we have people graduate from this institution who are practicing at the top of their license, whether they be a radiologist, a cardiologist, a nurse, a pharmacist, a public health official; how do we really embed the role of data and decision-making into the professionals for tomorrow and then heal is what we do at UCI health, right?

(06:19)

How are the patients that we’re seeing today getting the best possible care informed by all of the patients that we’ve seen before? You know, we have a tagline that the one hundredth and first patient at UCI health should have their care informed by the a hundred patients that we’ve seen with that disease before them. Isn’t that what insight and wisdom is really all about, is that we’re learning along the way and that every patient gets the benefit of the learning. That learning is experience of the doctor or the medical professional. But it’s also in the data that we’ve generated around all the care that we’ve delivered over time.

(06:53)

I like that. Something I’ve heard you talk about before in terms of data collection around a person, a patient, is our data you get from like your wearable health tracking devices. And we all can probably name one or two, and probably many of the people listening are wearing one right now.

(07:15)

Yeah, that’s one the things that we’re doing here at UCI and we’ve got a great technology and business partner called Syntropy that we work with. They have, they bring a data platform, right. And it doesn’t care what type of data. And so, one of the big shifts that we see, here at UCI we have the Institute for Integrative Health funded by the Samueli family. And, that really looks at health and wellness. That’s a huge part of what we’re doing here at UC Irvine, right? No one would ever grow up wanting to be a patient. And so what I believe is that we’re all fighting time to try not to become a patient. And so how do we build the empowered health consumer in our society?

(07:59)

Well, part of that is what’s the role of data? For those of us who might wear something like a Fitbit or an Oura ring, right? It generates data. It presents that data to you and helps you understand how you might make the best decisions, whether it’s more steps, get more sleep, eat a certain diet. And so that’s going to be a huge part. Now what’s interesting is that is something I get every day, which when paired with the once a year visit to see my doctor and that lipid panel combined together create different types of insights around my health that I can understand, and that I can talk to my doctor about as an advisor. Just like I have a financial advisor, my primary care physician should be a health advisor to me talking to me about what we could do to improve my health and my longevity and living a healthy, vibrant lifestyle. And that’s the kind of primary care physician that I have. We talk about my data. And I think this is a trend that you’ll see more and more because more and more of us are subscribing, whether it’s the Fitbit on our wrist or the Peloton device in our home, or whatever it might be, more data signals are being generated. And the more we collect them and connect them to other types of data, the more we might be able to generate insights for ourselves, but also for the medical professionals that we interact with.

(09:21)

So what I’m hearing is our medical professionals are working off really all the data that they have to make the best decisions for us, for patients, et cetera. But there’s probably a lot more data out there when you talk about the data we’re generating at home on our home devices and those kinds of things. Can you talk a little bit about that and how that could impact patient care in the future?

(09:43)

So, data in healthcare is growing exponentially at about 36% year over year, it’s exponential growth. The decision-making in the traditional healthcare environment, we use roughly five to 8% depending on whose study. So we’re using a small amount of the data available to us. Some of it is the data is not accessible in a usable format. The other is medical professionals are very stretched for time, right? So it’s also about the curation and presentation of the data that might make it more useful. And this is where tools and things like artificial intelligence, AI, could come to bear to bring more data together, synthesize it, make it presentable in a way that’s useful. When it comes to patients actually making their own healthcare decisions it is less than 1% of the data, which I find almost egregious, you know, as both someone involved in the healthcare industry and as a health consumer. I would love to be able to make decisions around my own health based on information.

(10:40)

If I was running a business and I said, you know, hey, you have to run this business based on 1% of the information about your business, how successful could I be as a business owner? Right. So how successful can I be in managing my health? So, yeah, so we have a lot of upside in health and healthcare to help our patients make better decisions for themselves, and for our clinicians to really understand their patients, the individual person that they’re dealing with. And that’s really what we’re doing here at UCI through the College of Health Sciences, UCI Health, and the new Precision Health Institute that we’re going to be launching very soon.

(11:14)

Wow. So, data’s only going to grow. It’s only just going to keep growing exponentially. So we’re never going to run out of data. It’s just a matter of curating it, presenting it, focusing it.

(11:26)

We have more data than we’ll ever know how to use. It really is. And this is where people talk about more data, you know, not 10,000 records, a hundred thousand, a million. It’s not about having more of the same type of data. It’s really about stitching together different types of data and contextualizing it to the problem statement that you’re identifying, which could be a research problem statement, or it could be, what are we going to do with the influx of COVID positive patients that are coming into our health system this week, which is what we dealt with here at UCI in the early part of 2021. Right.

(11:59)

Well, that’s a good transition to our final question. Actually, you may have already started to kind of answer it, which is what is your vision for using collaboratories in the future?

(12:09)

Yeah. I like to start with this, what research universities do. And, here’s what I’ve learned. Research universities, we create the future, which is a really cool thing to say to you when you’re talking to employees or, you know, donors and stakeholders. You think about it, we create knowledge and we disseminate knowledge. We educate the current and the next generation, right? So we create the future. It happens here first, which is a really, really cool concept to start with. So when I thought about what I could do here, when I came to UCI and you know, my expertise around technology and data is this concept of the collaboratory was how do we put our data to work, to do that right. To create the future. And so, you know, we’re doing this in a way that is somewhat unique because we were looking at the internal uses of data, but also the way to take that same data and make it an external asset to work with the community, whether it be the federal research agencies or industry, and then we’re wrapping around it, things that are really important to us, like inclusiveness, equity, our public mission to share.

(13:15)

And so if you look for example, the machine learning algorithm, we call it the COVID calculator that we generated, during the, the early 2021 surge, where we were actually finding out which patients were the most severe to help us figure out which patients we needed to get into bed, versus which patients we could send to do an outpatient infusion. We not only learned it, we fed it back to the federal government and informed them about more types of patients that could get these infusions done in an outpatient setting, so saving beds for everyone. And then we shared the model with the rest of the world, and we said, look, we’ve learned some things here. We’re a public university. We’re not holding this tight. We’re not considering it intellectual property. We want the rest of the world to have it, because if you can save one life by using the model that we developed, then we’re, we’re fulfilling our public mission.

(14:08)

So I think there’s a lot of things that we could do. The collaboratory is what I’ve learned since being here because 2020 was about starting the collaboratory in health and wellness. But as I started to learn more about UCI and talk to more, really, really innovative people here was that the collaboratory concept was a playbook that we could develop. We could take different domains and take this same approach towards data and build a collaboratory concept on the other things that we were doing. So the, the second one that came into existence is a collaboratory around student success, because we’re very interested in helping our students succeed, to closing the graduation gaps between different types of underrepresented groups. And we’re actually starting to use our data to develop those things. And that data also helps inform research that we’re doing with foundations and with groups out of Washington I’m really interested in. I started the conversation with a couple of our deans about how could we build a collaboratory under the concept of climate change and sustainability, lots of different types of data coming from different sources, that need to be brought together.

(15:14)

And one of the key things for me around the collaboratories is there should be an application part of it. You know, the nice thing about the collaboratory for health and wellness is the things that we develop. We put into practice at UCI Health, for the benefit of patients, what we’re doing in student success. We’re using those tools to help our patients succeed, help our advisors give the best advice to patients to help them get to graduation. The same thing is true around sustainability. We are one of the most sustainable campuses here at UCI. And if we had a data model underneath that in an engine of analytics to how do we do that the best. And we take that and we start exporting it out to help other universities and other communities be more sustainable. Then we have an amazing concept that we can replicate into different domains, where we see that we want to have a tremendous impact. So that’s what I see is more collaboratories, but specifically picked about places where we have deep expertise and also we’re going to apply what we’ve learned out of the data.

(16:16)

Yeah, you really touched on the main tenants of UCI. This is really exciting. I mean, it really has exponential possibilities.

(16:25)

I really think it does. And it’s exciting to come to work every day and have such a great community to work with.

(16:31)

That’s great. vice chancellor Andriola, thank you for speaking with me today.

(16:35)

Thank you so much for having me and I look forward to listening to all the different speakers you have on there. It’s been a great series. Thank you.

(16:43)

I look forward to hearing more about collaboratories in the future. For more information, you can visit the Office for Data and Information Technology website at odit.uci.edu/initiatives/collaboratory. UCI podcast is a production of Strategic Communications and Public Affairs at the University of California, Irvine. I’m Sheri Ledbetter. Thank you for listening.