In this podcast, Google Cloud’s Chris Haas, SpringML’s Sindhu Adini and Washington University in St. Louis’s Dr. Philip R.O. Payne discuss how a data-driven approach can help drive efficient campus reopening solutions for student and faculty wellness.
Speaker 1: On behalf of Google Cloud, SpringML, and Carahsoft, we would like to welcome you to today's podcast focused around campus reopening solutions for higher education were Google Cloud’s, Chris Haas, SpringML’s Sindhu Adini, and Washington University in St. Louis's Dr. Philip R.O., Payne will discuss how a data driven approach can help drive efficient campus reopening solutions for student and faculty wellness.
Sindhu Adini: So, you know, we've heard a lot about campuses, higher ed campuses, talking about laying out plans for reopening in fall 2021. Obviously, with the pandemic, you know, hitting us, a lot of campuses took different approaches, things that they've never done in the past, like some have taken a hybrid approach. And, you know, all the challenges that been faced by these campuses were, like, really first time challenges. So today, we're going to dive deep into what were the challenges, we saw a couple of universities like Washington University, who kind of took proactive measures and opening up campuses ahead of vaccine rollout. So let's talk a little more about what are the challenges and how Google answering ml enabled overcome some of these? That being said, Chris, in your kind of experience, and opinion and what you are hearing, what are some of the challenges that you saw this particular domain face? And what is you know, the value that it will bring in reopening up universities to some sense of normalcy?
Chris Haas: Yeah, thanks, Sindhu. And it's a pleasure to be here. Thank you everyone, for attending. We're given our large work with universities and K through 12, higher education customers, both in the US and throughout the world. When we've interviewed them, as the pandemic came on, we saw some large areas that I think we're all aware of. So just kind of reiterating, I think we all know to be true. And then we're coming out of out of some of the financial crisis for universities that's given that's that they are opening back up. I think they're things that Dr. Payne will maybe discuss around, you know, things that we didn't necessarily anticipate, like positive things that came out of the pandemic and having a remote learning. We know already, there's a large financial impact against all universities, just lower enrollments, not no one on campus. So there's obviously financial impact there. More importantly, big impact on equity. We know that, you know, I see this, and my young kids have told me, my younger kids school, but also universities, how minority populations are more highly affected when it comes to access to equipment and high speed internet for remote learning. So it's a very big impact and obviously, causes a loss of minority populations, maybe even going into universities. And then snowing that because there's this remote learning need or has been, you know, we're coming out of that now. But for this, this current school year, large remote learning need, students are going to look for schools that are differentiated ones that offer better online experiences. So that's going to cause an impact if you weren't able to pivot quickly enough to that remote learning. Luckily, I know a number of you are leveraging more cloud based solutions. So might not might have been easier to pivot into this remote learning. But other universities, and K through 12, obviously could struggle in that area. And then finally, just the overall student experience, we know that everyone wants to have not just the academic experience in universities, but they want to have that relationship building experience being out on campus. And without that, it's going to drive down enrollment. So it's important to have the technology to back all of their learning needs, and then also enable that campus reopening, leveraging similar technologies to make sure the campus is safe. And then just the local economy, economic impact is huge of having a campus open anything from those students working in the local economy, to shopping to go into events, restaurants, eat out conferences, traveling to and from that local area. So you know, impact on the local travel as well. It's going to have a large impact financially on the local economy when the campuses reopen. So it's a goal that we're all striving for. I know we're moving very quickly, you all responded in incredible ways during the pandemic and we know this to be true that having a campus open will make the local economy better.
Sindhu Adini: You had a great point, Chris, because you know, when we look at our kids who are from k to 12, we clearly see you know, how the pandemic has hit and kind of widen the experience in terms of equity. But Dr. Payne given that you know, via hearing about all this, what was your experience when the pandemic hit your, you know, your university specifically?
Dr. Philip R.O. Payne: Yeah, so it's a great question. You know, it was sort of a dramatic turn of events at the beginning of the pandemic, in a period of just a few weeks, we went from, you know, asking the question of whether we needed a plan to respond to COVID-19, to moving to a situation which we had no students on campus, and we had shut down all in person instruction, research and non-essential clinical activities in our medical center. And it happened very quickly. And then the question sort of came up what next, and I think that we quickly ascertain that there were a couple essential aspects of deciding what that what next would look like. The first was that we were going to take a data and scientifically anchored approach to making those decisions. Probably not accidental, given that the chancellor of our university is a quantitative political scientist. And we have one of the nation's largest academic health centers, and a leader in this area, including infectious diseases and global pandemics. I think the second thing we recognize is that we needed to the point that Chris just raised, we needed to find a way to bring as many of our students back safely, and provide alternative mechanisms for our students to engage in education and research if they couldn't come back to campus. So really a hybrid approach and, and leading with that hybrid approach. And then I think the third key dimension, again, Chris, sort of alluded to this was that we also knew that we had to continue to serve the community that we're located in, you know, Washington University has, as part of its emerging strategic planning process at present with our new chancellor, focused on what we call the St. Louis initiative, which is predicated on this statement of being in St. Louis for St. Louis, rather than being focused solely on what happens in the confines of our campus. And that really meant that from the outset, we were not only worried about how we took care of our students and our faculty and staff during the pandemic, but also how do we serve the needs of the community and to the extent possible, mitigate some of the impact of the pandemic, because we are a major driver of economic development, employment, purchase services in a variety of other sort of positive economic sort of activities. And so we knew we had to move quickly and be proactive, as you alluded to earlier, simply right? We had to make those decisions, not in a matter of months, but really in days and weeks. So we could figure out what that new normal would look like. And that's really how we approach it from the start. And it's how we approach it today.
Sindhu Adini: Yeah, and I think you hit a key point as to what decisions do we make? And how do we know it's the right decision, and taking kind of that data driven approach is something that we'll discuss as we go through, you know, what was done at Washington University. But before we dive deep into that, Chris, you know, we spoke about the seminar is about, hey, how did Google and string ml kind of come together? And can you talk a little bit more about, you know, the breadth of what we've done in the space, not just from a university perspective, but from a community perspective, and then kind of narrow down on hey, what made sense to ensure that universities were safely reopening the campuses?
Chris Haas: Sure. So first, I just wanna acknowledge universities are unique. And when it comes to working with local and state government, every commercial entity company does as well. But universities have, like Dr. Payne just said, most universities have some sort of alignment with public sector from when it comes to research. And I'm using that in a broad sense of things not just related to COVID, just broad research. And we have focused our public Google Cloud public sector team along with partners like SpringML, and our customers, and partners, like Wash U, have really focused on developing solutions that aid both in responding to in this case, specifically, the pandemic. So early on in the pandemic, we started seeing the need just to do undiagnosed symptom tracking, that was the first solution that we brought to market. And many of the example customers, we just saw on the prior slide have adopted that plus other solutions that we've had, you most University have also taken on simple tracking, contact tracing, exposure notifications, just as some examples, so that we've built this breadth of solutions to really aid you in an end to end response to the pandemic. So starting with things like I've already mentioned, symptom tracking. So undiagnosed students be able to self or self-respond to the University of hear the symptoms I'm having today, and I'm feeling healthy. So I can come back to that need to do contact tracing. And we'll talk more about that both active tracing, and then really not contact tracing, but a different lens using exposure notifications. Underlying all of this, right, you're gathering lots of rich, important data from students from faculty from activities on the campus. So unlike everything that I've just discussed, is this data analytics platform that lets you do really robust analyses like density tracking, compliance, just understanding like what people are badging in or coming into two buildings. And did we receive a health check that day from them or not? Were they registered or not. And then enabling both the online experience. But this omni-channel experience through intelligent virtual agents, now intelligent virtual agent could be online, just something you could interact with to really find out or submit your health check. But also enabling via phone, because we know there are lots of people that aren't necessarily able to or wanting to interact online. And they want to call and frontline in a call center through an intelligent virtual agent is a very powerful way to scale the university's capabilities when it comes to having a safe open campus. And then finally, just immunization credentials. This is a kind of a TBD thing. Everyone's thinking about it, but it's going to be important. I don't know exactly how it's going to come out yet. There's no you know, there are standards bodies that are working on it. There's lots of organizations that are engaged with this globally, there's going to be need for it I and I just think it's something that we all need to be aware of, because having disjointed solutions for immunization credentials is not going to be a thing that's going to benefit us in the future. So coming up with this concept of a immunization credential, to understand students in this case, coming into a campus and ensuring that they are, you know, safe to return to class.
Sindhu Adini: I know one thing I think the pandemic has brought to us is, you know, all these capabilities are great. And in a typical pre pandemic world, we would have done this in our own pace had like entire this thing. And this would have been something that we would do in an innovation lab or something like that. Chris, can you talk a little bit about how Google and SpringML kind of did this rapidly. And then Dr. Payne Can you also talk about the experience and actually understanding the need and agility and the speed at which this had to be deployed in your experience in standing something up for your university?
Chris Haas: Yeah, that's briefly targets, like I hand it over to you, Dr. Payne. Google Cloud SpringML, we are designed to move fast, we're very agile, I mean that both in front project management perspective, our approach is very agile methodology, but also just agile, meaning move fast pivot as needed quickly, that is really designed both in our how we approach implementation of solutions with customers, our partners yourselves, but it also is how we build our technology. So we have a heavy focus in this platform as a service or serverless realm of technologies, meaning you don't have to worry about the underlying infrastructure, like you wouldn't in the past with more virtualized infrastructure, you can really worry about just delivering the solution, gaining the insight you need from it, and the infrastructure is just there and ready to scale as needed. Because of that, you have less need from a technical staff perspective, because you're having to patch operating systems, you're not having to install applications. On top of that, you can really focus again, on that business logic layer. And not all the underlying aspects that deliver that business logic and flexibility, our solutions can be very lightweight, to more robust, and you can adapt, adopt them in a spectrum from lightweight to more robust as you need. So adding functionality, starting with something smaller that delivers on a prime need, and then adding to it as you have other things that you need to deploy to enable functionality to make sure that you can open the campus safely.
Dr. Philip R.O. Payne: Yeah, so maybe I could just add to what Chris said, an interesting experience here at Washington University and working with our partners at BJC. Healthcare, which is our delivery network that partners with our School of Medicine, early in the pandemic. And I mean, in the first few weeks, when we recognize what we were facing a combination of the Chancellor of the University, our Executive Vice Chancellor for medical affairs, and the CEO of the health system, as well as his leadership team came to a unit within the university called the Institute for Informatics, which is one of the units I'm responsible for. And we had built the Institute for Informatics to behave sort of like a startup inside the university really more agile and entrepreneurial than a traditional academic unit. And that was something that had been done intentionally. And they came to us because they realized that some of the really pressing problems that we needed to tackle in order to create this situational awareness. So we could, as I said, before, you know, bring back as many of our students as possible, operate our health system serve the needs of the community, there was a recognition that we couldn't approach those problems in a traditional way that takes, you know, months or years, all of a sudden, sort of the timeline for success became days and weeks. And so we knew we needed a differentiated approach to deploying those technologies. And so that's where our partnership with Google and spring ml became so important. We had an existing relationship with Google Cloud in particular as it related to scientific computing. And we really capitalized on that relationship to say, how quickly can we move? How quickly can we build a platform in the cloud, along with mobile applications, web applications, any number of other endpoints that allow us to bring together contact tracing, you know, daily symptom monitoring, security data, local and regional epidemiologic data and I can keep on going down the list so that our decision makers could look at that and make critical decisions in real time. And I'll give you a concrete example To that end, you know, early on before we had deployed the platforms that we were able to develop working with spring ml and Google, we had a question raise, which was we had a prototype of a daily symptom screening application. And we'll ignore the fact that it was buggy and didn't scale particularly well. But luckily, we only had, you know, a few 1000 people on campus at that point in time. But we had the question rates, could we link that data from that daily symptom screening application, with card swipe data from our buildings, or the plans that said, who was authorized to be on campus at that point in time, given some of the state of emergency that we were navigating, and there was an initial version, that report that didn't work particularly well put together and it took close to a month to put that together? While we were simultaneously building the platform out with spring ml Google, as soon as the spring event on Google Platform went online, we went back and re-engineered that report, and we did it in about two or three days. And we use that report every day now to find out from a compliance standpoint, who's not completing their daily symptom monitoring questionnaire? And are they swiping into buildings? And where are they? And you know, what are they authorized to do? And how does that relate potentially, to both our testing and contact tracing programs, if that's necessary? And I point that out, because, you know, in the past, that's the sort of project that a university you know, between its analytics, and it groups might take several months to implement. And we literally did it in a few days working with spring, ml and Google. And I think that speaks to the types of agility that we need. I think the problem actually, and I mean this in a positive ways, we've now become a little bit accustomed to working like this. And the question is, how do we sustain this beyond the scope of an emergency like COVID-19? And I will say, before, I have sort of wrap up this set of remarks that we've already started talking about, how does the platform that we've been building with spring ml and google translate into something that actually supports the longer term health and wellness of our campus community beyond COVID-19? Because there will be a time when COVID-19 is largely treated as an endemic disease, just like we treat seasonal flu, but there will be additional, you know, challenges, and how are we going to navigate that, and I think these tools will be just as instrumental to those, you know, emergent health and wellness challenges as they are in the current emergency situation. So agility and speed and to the breaking of that normal sort of currency or timeline of success being months or years. I think it's a whole different ballgame. And we need these platform as a service capabilities and our partners outside the university to move that quickly.
Sindhu Adini: Yeah, and I think you hit on a great point there, Dr. Payne is that we talk about agility and flexibility. And when we talk about, you know, cloud services, we talk about, hey, here is the scale in which, you know, a component can actually scale for vertically or horizontally. But I think one key aspect that has come out of building these applications and during the pandemic is, how can this scale for additional use cases like you don't have to rip and replace? How do you use that foundation, and keep building to solve additional use cases, we always talk about how data is, you know, the new oil or the New Sun, you can reuse it for additional use cases. But looking at using platform as a service? Also, how do you reuse that foundation for additional use cases just beyond what was built for COVID? Let's narrow down a little on, okay, as we went into this pandemic, and we wanted to ensure that we open up the campus safely, what were the Chris, what were the key pillars, that, you know, we thought of when we wanted to, you know, work on a approach to allow students to safely return to the campus?
Chris Haas: Yeah, first I'll pivot on what we were talking about the horizontal platform capabilities and adding on to that. So think of this like you, we have these columns of areas where university administration is wanting to gather information from so they can make more data driven decisions underlying this, if you think of this as like a logical architecture, you have this horizontal cloud platform that can really do anything you want it to do without the limits of traditional hardware and traditional data centers, meaning you're not going to run out of capacity, you don't necessarily you do have to have capacity planning, but not in the same vein, you can really just say, I have a platform. Think of it this is just we just focused on the data analytics side of this, right? You're gathering information rich information from every one of these pillars and delivering it back out. But you're putting it in this horizontal platform, that's just scaling and gathering the information you need. So you can do those quick analyses like Dr. Payne said, being able to pivot quickly having data in a centralized location that you don't have to worry about the storage limitations of it. And then being able to within days build a situational awareness dashboard that gives you the insights you need to then make more informed decisions on campus reopening and the well-being of the students and faculty on the campus. So thinking first about student and campus health. So I've mentioned symptom tracking and contact tracing, and exposure notifications. Dr. Payne mentioned security compliance badging in all of that information, build a gathering that in real time as that data maybe, but ingesting it so that it's there and able to be analyzed, when it lands in the platform, from all those various sources, right. And I'm not even, I mean, the various sources, we could talk all day long, just on where you could gather data from, it's not just from the solutions that are being deployed for response to COVID. It's all those public data sources that are available, be able to ingest that data into the platform as well, to make those operational decisions, like who should be returning to campus, what wave should we implement, to align with the student body that we and the faculty body that we have coming back onto campus, as well as everyone else supporting the campus, and then just ongoing analysis of the safety of the campus, right, because you again, the state is all in the platform, you own it, you choose what to do with it, you choose how to retain it, but it's there. So you can just do this ongoing analysis of trends over time, as just a simple example, density tracking is another one as another example. And then the communication platform is really critical. You have all this rich information, you can use that to build FAQs and knowledge bases, that then is used to power an intelligent virtual agent, to be able to be responsive to student family faculty needs when they have questions around who is able to return now? Are they eligible? How do I submit that I'm healthy that day? And do that through this omni-channel communication platform?
Sindhu Adini: Thank you, Chris. And Dr. Payne, like, as we spoke through these problems, did they resonate with what your team was seeing at the onset of the pandemic? Were they problems that were like, you know, more challenging to solve before we had this platform in place? And where the problems that you know, that came from your blind side that we didn't even anticipate? And how did how was the team or the platform able to kind of react to something like that saying, Hey, we didn't even expect that that could be a problem?
Dr. Philip R.O. Payne: Yeah, so it's a great question. So I would start by saying that I am prone to describe the environment at Washington University, including our technology and data environments, as being pathologically decentralized. And as a result of that, you know, the data assets and the tools that we needed to make sense of that data in response to the pandemic. We're very much distributed across the campus in a variety of silos. And what we quickly ascertained when our leadership was saying, you know, you know, how many students can we bring back and, you know, have an adequate amount of, you know, housing, including isolation, housing, what volume and frequency of testing is required in order to manage the campus committee and make sure that we identify, you know, sort of potential outbreaks, if there is an outbreak, how will we sort of understand from a, you know, sort of epidemiologic standpoint, where it started and how it might propagate across the campus. And I could keep on going down this list. But all of these involves sort of connecting the dots across a variety of data assets on the campus. And I think that the real blind spot that we identified is that because of that, as I affectionately referred to it, pathologic decentralization, we didn't really have the connective tissue, you can tell I'm a faculty member in medicine, so I had to keep on using all these healthcare paradigms. But we didn't have the connective tissue to bring those data together quickly and easily. And every time we asked the question required all these sort of ad hoc and very bespoke sort of hand coded solutions to bridge those data, and that just wasn't fast enough, or reliable enough to meet our information needs, which goes to the point that Chris was making earlier, right? We really needed this platform where we could stop worrying about how is the data stored and how is it accessed and the sources as we can sort of deliver that data alongside new data sources, like we've already talked about, such as symptom screening, and contact tracing and the like, and then produce measures and visualizations and reports, you know, in a matter of hours, and, you know, we simply did not have a shared platform, and a, you know, sort of ability to connect those dots across all of those capabilities. And I will tell you that, you know, early on, we had some very tough conversations, and people said, why is this so hard? And the answer was, because we've never taken the time to, you know, really think about how all of these individual data assets on our campus, connect with one another. And then, you know, the next question that we were often asked, especially by, you know, sort of our senior leaders was, you know, why is it that people in private industry or in other economic sectors can do this and we can't, and the reality is, because in other economic sectors, there has been more proactive investment in the types of data platforms that enable situational awareness, which has not been as high a priority in you know, higher ed. And so when you put that all together are you know, a real blind spot. And that was it was a blind spot for some of us, but maybe not all of us was the lack of that sort of ability to connect the dots quickly and easily in sort of a robust way. And so that's where the work that was in front of us became so critical, it wasn't necessarily that we didn't have the data in its source systems to understand what was going on, we should know how to get to that data and combine it and then deliver it in a meaningful and actionable format. And when you're starting to make decisions very quickly, that you know, quite honestly have real life or death consequences. Because, you know, we talked about the fact that we needed to return our students to campus in equal measure, I needed to have a plan to bring back our clinical faculty and staff, and clinical trainees to run one of the country's largest academic health centers that ultimately took responsibility for almost 70% of all ICU admissions for COVID in a 300 mile catchment area. So in equal measure, to taking care of my student population, I need to do all the things we're talking about to make that health system run. And so you know, they're the stakes are quite high. And again, we had to build to connect the dots to be able to make smart decisions very quickly to respond to what was a very dynamic situation. And, you know, we look back at it now, a year later, and we say, okay, you know, we're really glad to sort of, you know, have a little bit more understanding of the work that needs to be done. I got to tell you at the time, you know, we were having critical conversations, like, can we even, you know, get enough reagents to test people? Do we have enough PP? What's the potential mortality rate? If we bring too many people back to our campus? Right? Those were the sorts of conversations we had to have early on, and they all require data to actually, you know, really be able to ask and answer those questions.
Sindhu Adini: Yeah. And I think that's a great segue into, yes, we spoke about, hey, here's all the data, the silo data, how do we connect them to make those insights and those decisions? But then again, how do we actually approach this problem of bringing the data together and making the right decisions but at a scale? How do we ensure that what we are solving for also solves for, you know, the student journey? Because when we actually, typically as technologists, when we start looking at your is an insight that I need to get we start looking at what are the data points, the technology and the scale for, you know, hey, how do I scale this technology to bring in the data points. But Chris, can you talk about how we took the approach to scale it for something like a huge University, how we accounted the user, how we kind of accounted the user journey, and how we enabled this functional aspect within Google Cloud?
Chris Haas: Yeah, first, I have to say, Please don't take this the wrong way. I mean, it in a positive light. I think public sector universities, you're extremely data rich, your intelligence, which means you have you have these the individuals who literally develop these technologies that then the world adopts. But because you were the earliest adopters of technology before commercial, right public sector, government and universities adopting technology decades before commercial entities really were, that's your technology poor or have been technology poor traditionally, because you've had to adopt these more point solutions over time, which means you have to maintain them over time. So pivoting to the cloud, and leveraging the cloud to can help you leapfrog really, in my opinion, over where maybe a commercial entity is right now, cloud gives you the ability to really leapfrog So as an example, leveraging something like a cloud based, it could be a more installable mobile application, or could just be a web based mobile application that that students and faculty use. This also includes that omni-channel experience I'm talking about as well. So not leaving out people who, who don't want to or can't access this functionality online. But having that capability in the hands of the students and faculty to then give that information daily back to university will start building this aggregate level of datasets that they can leverage. Now talking about how you can integrate with as needed local testing, and other hospitals that might not be part of the actual university system, but are serving the needs of the faculty and have the students be able to integrate with them, leveraging cloud based technology, across a hybrid cloud environment, as well as the concept of the edge. The edge in this case could be the actual pharmacy delivering tests, or the hot the other hospital entity delivering tests, and be able to feed that information back into the university's platform, building all this aggregate level data set that you can then analyze and then respond as needed back to all of the stakeholders, both the administration but also the families of the students and the faculty themselves. As far as you know, what's the latest information coming out from the CDC? What's the latest guidance from them? What's the guidance from the university? What are the parameters by which you can come onto campus safely and health healthy? And just how can you continue that bi directional communication with them to make sure they're feeling informed and safe when they're returning to school? And that's talking about you're taking that just to down to the user journey and the fact of, you know, in this example, how having an open campus with that wellness application the student can leverage enables them to really, you know, eventually be more safe in doing things in the local economy. So we're going in this example just going to an event, and being able to do check ins and making sure that they are healthy first to go. And then once they've done that, be able to flow through the whole journey to go through that event, and then still return to campus at the end. And just having these horizontal technologies underneath that really enabled us. I've mentioned intelligent virtual agents, our contact center AI and dialogue flow technologies, the University leveraging Chromebooks to do badging check ins, things like that, to return to campus, and communications outside of just the core, I mentioned already with intelligent virtual agents and the call center, but leveraging things like Google Analytics on the websites to understand what people are doing when they're coming to your websites, publishing videos around safe campus reopening on YouTube. And then ultimately, again, you have this broad data set underlying everything we're doing here, but be able to build those situational awareness dashboards so that everyone in the administration is informed understanding what's happening with the campus, how safe it is, who's returning, and how the campuses operating, leveraging various visualization technologies that we have.
Sindhu Adini: Thanks, Chris. And before we go into scale, and you know, some details around tracing and monitoring. Dr. Payne, what have you heard from your students and staff with the solution? Right? Like, yes, it solves for all this, we're talking about this journey? Is this really hypothesis? Or is it really kind of, you know, what you're seeing in reality with your students?
Dr. Philip R.O. Payne: No, I mean, I think this is exactly what we've been seeing with our population, especially as we brought more and more of our students back to campus. You know, when I think the remarkable things about this is, you know, this journey that you see sort of charted out here, you know, in a prototypical, you know, archetype. You know, it involves a lot of new uses of technology that we've not seen on the campus before, right, you know, whether that be the symptom screening apps or tools for checking in and getting your screening tests or other capabilities allow people to make smarter decisions about travel, or you know, any number of other dimensions of sort of this new normal that we're living in. And the thing that I want to emphasize is that through the use of sort of the, you know, responsive, you know, web apps, mobile solutions and other tools that we've been able to develop working with spring, ml, and Google, these have become very embedded in our campus culture, right, this has become sort of the norm. And, you know, our students are actually quite used to it now. And except that this is part of the campus experience. So I think in many ways, one of the things that we've done is sort of re-engineer the culture of the campus experience. And in a way we've put health and wellness front and center in a way that it never has been before, right, we're asking every one of our students every day, how are you feeling? Right now, we're asking targeted questions related to potential COVID-19 symptoms. But more broadly, we are asking them to provide feedback, where if important signals are found, it gets routed to the team in our Student Health Center, and they will follow up and determine whether those students need additional support from a health and wellness standpoint, we've never done that before, right? Similarly, we are creating sort of a shared compact around safety in the classroom or in other shared environments by saying, you know, you need to complete these screenings and have your daily screening app and actually, you know, check in with, you know, the people managing that facility or location. So they know that you have completed that, right. And we have monitoring compliance to make sure that people are sort of, you know, adhering to that sort of social compact. So again, we create the social compact, where not only are we looking at the individual level, but we're also looking at sort of the collective level. And there's this expectation. And you know, there's interesting byproducts, which I won't belabor such as the fact that we had, famously a Instagram account that popped up where people were posting pictures of individuals not adhering to masking mandates. Wouldn’t say that that was a university endorsed activity, but an interesting use of social media to create that social compact. But the reality is, these two become embedded in the campus community. And I would say that, you know, what we've really done is we've re-engineered the campus journey. And I often tell people, and I know it sounds a bit metaphoric, but you know, when people talk about what comes next, it's not a return to where we were before this pandemic, because there were some aspects of pre pandemic campus life that certainly we want to return to. But in equal measure, there were some aspects of pre pandemic campus life that didn't adequately reflect concerns around health and wellness and sort of collective responsibility for safety on the campus. And we want to keep those dimensions as we move into what I always refer to as a new normal. So you know, I always tell people, we're not returning normal. We're gonna engineer what comes next based on what we've learned in the last year and this campus journey here. I think there are dimensions of it that will continue right. Whether it be you and I didn't talk about a transportation or food or you know, social engagement on the campus, which is also fundamentally changed. So I think it's really important to recognize we're engineering in real time a new student experience as a result of what's happened in the last 12 plus months. Great. I mean, that's good to hear always, like, you know, on paper, it sounds fantastic. But to actually hear from you that this is how the students have been. And they're kind of leveraging the new technology embracing it. And you know, now it's probably a part of their regular normal life, or the new normal as you say it just for a moment, eleventy for this group, something I mean, we talk about the new normal, one of the things that always strikes me is we have a marketing campaign around the daily symptom screener, and we have signs everywhere on the campus. And like literally 1000s of these signs with different you know, sort of calls to action to make sure you complete your daily symptom screener, it's got a QR code to make sure you can get through it. And I have a tendency on the weekends to walk my dog through the campus. And like, every once a while, I have to like step back and say, Oh, wait, that project with the 1000s of signs on the campus. That's my project. You know, it's interesting, but I mean, it's everywhere, right? We've embedded it, you know, there are social media campaigns, there's physical media, there's all these other steps, you know, it's become endemic, but it's like I described the virus, these sort of daily use of these technologies become endemic on the campus.
Sindhu Adini: Yeah. And I'm sure then we actually even dive into analytics, you can tell us more about how you've measured the success of those marketing campaigns. And before we dive deep into that, Chris, would you tell us how we functionally enable this? And for our audience here, who has different views of what digital contact tracing is, what are the different types that we do here? That, you know, we know about in terms of contact tracing? And also, we'll quickly pivot to, you know, Dr. Payne to talk about what, what is the type of contact tracing that we leverage at Washington University, and what are the other additional aspects we added to it, and then look at how you're actually looking at the data and ensuring that and the patterns and the trends you'd see from the data being gathered.
Chris Haas: First, I want to comment, and we could dive into this, I don't mean to take us down a rabbit hole. But just picking out what Dr. Payne said, it'll be interesting to see how you take away the need to gather specific information from students and faculty around COVID-19, to the future of just general wellness. And now that we've gone through this digital transformation, and we kind of have a level set of behavior and acceptance of, you know, submitting information so that you can be more better taken care of how maybe something like these applications can be changed into something to just general wellness and collection of wellness information and health of individuals not necessarily identifiable. I agree level just understanding is my campus feeling well, in general, outside of being sick. Now I'm going to talk about sickness, it'd be interesting how maybe these technologies could be pivoted to that in the future, just understand wellness generally, which I think is something that is probably going to be well has been on everyone's mind for a long time now. But I think one of the features or probably even more so. So anyway, back to the slides here. And I've talked a lot about this. So I'm trying to, I'll try to pick it up. But again, the whole idea is that you're deploying applications quickly, to respond to the pandemic, in order to safely reopen campus. So symptom tracking, contact tracing, many other examples, communication platform, generating data that can be routed immediately to some sort of support service, right on campus support service that they can take action on it. But also falling that data into this analytics platform so that it can be aggregated and analyzed in real time, historically, however needed and then visualize to be able to make those data driven decisions. And that all being put into this smart analytics platform, as well as the underlying application, modern application platform that really powers everything we're talking about here. And just enabling the cyclical data flow, and data analysis and application scalability that the cloud platform delivers for the university and then serving the students and the faculty, the administration.
Sindhu Adini: Okay, let's talk a little about those types of tracing. Chris, and then we'll quickly you know, as Dr. Payne to describe what we've done for Wash U specifically.
Chris Haas: Yeah, and so just in case you weren't really aware, that aware of exposure notifications, which is technology developed by Google and Apple earlier last year, really, it's not trying to replace contact tracing your traditional infectious disease, contact tracing, epidemiological contact tracing, is still very much needed. We know that people are not necessarily going to respond as proactively to a kind of passive tracing using exposure of vacations as they would from an outbound call from a contact tracer. So there's still a very large need for active tracing. The problem that you have everyone has is scalability. You can't put enough people in seats to make all the calls you need to do that contact tracing. I think Dr. Payne for you the numbers, I think the average infection for flu was like two to three people beyond the infected person, I think COVID is triple that somewhere in that realm. So you can imagine just the need to do your how many people you need to do effective active contact tracing. But we have the technology to do that to enable the capability to automate that in ways that let you scale those contact tracers. And then you have the alternative of exposure notifications, which really lets the person being the control state of do, I want to notify everyone that I maybe even I don't know about, but I've been in close proximity with that I've been infected and share that back also with the state's public health authority and the university. And just again, you know, this whole cyclical loop of gathering information, analyzing it, collecting it, put it all together from the various sources, and then enabling those human tracers on the active tracing side to be more informed when they are doing that outreach, either through an automated, SMS or, or intelligent agent call out or actually in person calling out to collect the information by enabling the automated collection of the information being shared back as part of that contact tracing approach, and exposure modifications plays into this as well. It's not going to give the as rich maybe of information as active contact tracing, but it gives you the signals that you can then use to be more informed and how you empower the contact tracers and who they should communicate with first, and active monitoring. Again, it's just a more automated way to enable contact tracing. So through mobile applications, being able to automate the act of collection, automated collection of contact trace information could be symptom tracking, it could be tests, it could be a collection of who they've potentially been in close contact with, all through a modern platform that enables the actual students, faculty families to be easier and gathering information and providing it back to the university and the state public health departments.
Sindhu Adini: Thank you, Chris. Dr. Payne, can you like double click into how this translated for Wash U. You mentioned, you know, it was even beyond just the students? Can you talk a little more about the active kind of symptom tracking and contact tracing and exposure notifications?
Dr. Philip R.O. Payne: Yeah. So first of all, I want to go back to a point that Chris made, you know, for us, it's somewhere between 10 to 11 individuals for every positive COVID case. So actually, I think we've learned that your multiplier about three times is exactly right, with the challenge being that the sort of timeframe in which we have to identify those individuals and get them into isolation is much more challenging than it is with influenza. So that problem is, you know, multi-dimensional, and its complexity, to say the least, we started out our sort of journey in this regard, focusing on daily symptom screening. And we started first with our Health Sciences Campus and then expanded to the rest of our campus. And you know, this is delivered to, you know, all of our workforce as well as now our students, and we require this screen to be completed before people come on campus, we actually have a variety of entry points where it sort of screening results are confirmed. And actually we use a touch free thermal camera system to screen people further for their temperatures as well, a big part of this was actually creating a feed from this system in just a dashboard that would allow us to understand, you know, where are we seeing clusters of symptoms, right, you know, by what departments locations, what symptoms are triggering, we actually can drill down because if somebody gets a alert that says, Do not come to campus, or contact occupational or student health as a result of some symptoms that they've self-reported, we then go on to have them call those individuals and those individual actually see what their response was and confirm that with the person calling in and they may be referred for testing or otherwise told to isolate. You know, another big part of that has been a recently rolled out has been a kiosk solution for our contractors and visitors. And I think that's something that's very much under appreciated. You know, most universities have a high number of contractors coming on campus every day, whether that be facilities, food services, landscaping, construction, information, technology, I can keep on going on the list. And you know, they are not our employees, but they are in our campus environment. And therefore, we needed a tool to also help screen them and be able to refer them for subsequent testing and evaluation if they had, you know, potentially raise some symptoms. And I didn't say this out loud. But one of the things we did early on is we took responsibility for case investigation and contact tracing for all of our campus community, working in conjunction with our public health department. So we basically said to the public health department in both St. Louis city and county, if we had individuals that required isolation, testing or quarantine, we would be responsible for that. And that allows us to manage our workforce and be more agile and responsive to those needs. And so here you just see a few screenshots from those tools. And then from there, we obviously are now moving on to the deployment of an exposure notification tool for our community. And interestingly enough, while it's been developed for our campus community, there's already been an extreme level of interest from, again, our county public health department about making this more broadly available. And part of the reason for this is, it goes back to what Chris said. So we certainly have a robust contact tracing operation, both for Occupational Health and for student health. And we're actually using a tool that spring ml and Google have helped us to develop to manage data from those contact tracing investigations. But it's a slow and laborious process. And the reality is, is that can take too long to identify in a campus of our size, all of those cases that need to isolate or require testing if we want to slow down the spread of covid-19. And that's actually where the exposure notification tool allows us to basically let people know as soon as the positive test is recorded by occupational student health, that they may have been exposed, and that they should make the decision to isolate and with a call to action to allow them to directly connect to the appropriate occupational or Student Health Group. That's speeding it up before our contact ratios can get there. And we believe that this is probably the most important dimension of exposure modification is just shortening that time between positive test result of a close contact, and the decision to isolate and pursue testing. And this is something that we continue to work on, because that number that Chris alluded to earlier, because we won't have an infinite number of people to do case investigation. So we do have to triage and prioritize case investigations. And this is an important supplement to that traditional case investigation, I think the two of them working in concert is central to our ability to you know, protect the safety of our campus community. And we've been very fortunate, I will say from the outset, through all the tools that we've talked about in this webinar, we've seen an extremely low level of, you know, transmission dynamics on our campus. And you know, interestingly, we built models early on saying these are the worst case scenarios. And I'm very happy to have been proven wrong about all of those worst case scenarios, because we've come nowhere near that in terms of COVID transmission on the campus. And I think it's because of all the aggressive measures in data driven decision making that we just talked about.
Sindhu Adini: And the perfect segue. So Dr. Payne, can you talk a little bit more about what are those insights you saw?
Dr. Philip R.O. Payne: So you know, this is quite interesting, right? So you know, the background picture actually shows us a set of visualizations and sort of a dashboard that allows us to understand in real time, how many screenings that happened today, how many people have screened, positive or negative regarding the warning sign? Where are they located? Are we seeing trends in positivity or negativity in terms of, you know, potentially worrisome symptoms, and you can actually see sort of the sort of high daily submission alerts is interesting, dynamic, you know, what happens if we have someone that tries to submit multiple screenings, you know, in a shorter time period, you know, one of my favorite metrics, just truth be told him, we don't use this generally, but one of my favorite metrics is time between card swipe, and a parking garage to completion of the screening tool, you know, spoiler alert, we have a lot of people completing their screening tool in their cars before they walk into the buildings. So you know, we are able to find these patterns in terms of screening behavior, and then we can educate because I think that's the key aspect of this, you know, one of the things we learned was that we were having a lot of people's completing the screening out very close into when they entered the building. And we were asking people to take their temperatures, so the likelihood that they had a thermometer in their car was probably pretty low, back to the example I gave before. So now we've educated we have a campaign, make sure we literally send a message, you take your temperature at home, before you leave to come to the office or to the campus. So these trends not only are about identifying clusters or protect potential signals the environment, but they also help us to target our outreach and engagement campaigns to make sure the tools are used to the optimal effect in terms of increasing patient safety. And I can keep on going down the list of all the great things that we've found there. But I will tell you that you know, our ability to get these dashboards and reports in an easily digestible format. This is what our senior leaders need, right, they need to be able to click on a dashboard, and we just provide them all have access to it, they can see it in real time, getting that pulse of what's going on in the campus. That's it, they need to be comfortable with the pace of decision making that we've had to engage in the last 12 months.
Sindhu Adini: Thank you, Dr. Payne, and we'll send them your contact information. So you can talk about more of these patterns and insights that you're seeing and enlighten us and in the interest of time, we wanted to talk about what's next with vaccines rolling out and this is all pre vaccines with vaccines rolling out bring ml is working on this student digital immunization pass and in the interest of time without going too much into details of what the solution looks like basically ensuring that we are able to pivot to any kind of standards that come out. We know a lot of organizations are working on standards, ensuring that we are able to tap into the right data sources and ensuring that we are giving a more enhanced journey post pandemic and you know that this vaccines rollout, but we wanted to get your thoughts on, how do you see this fit in or the value prop of having an immunization credential, not only for, you know, the COVID-19 vaccines, but even beyond.
Dr. Philip R.O. Payne: If I could just jump in on this topic very, very quickly. You know, this is an essential problem for us in our campus communities, because the realities is as we bring more and more of our campus members back, right, because I talked about how we keep on bringing people back. But many of our international students are not back, many of our students who live at a substantial distance from the campus are not back, they're going to school online, we still have a substantial component of our workforce working remotely, because they are in positions that don't require them to be on campus. But we'd like to come back for interactive or other, you know, purposes, people are receiving vaccines everywhere. And you and I have talked about this, you know, and another part of my job, I'm responsible for sort of our data strategy around our community vaccine campaign, people are getting their vaccines everywhere, we have a variety of vaccine registries that are sort of maintained at the state level, working with the CDC, we have a variety of testing sources, which is a correlate set of data. And we don't have a trusted or authoritative source of that information that's quickly and easily accessible. When we are assessing again, the ability people come back to campus. And there are a lot of ethical and legal and social and other dimensions to implementing this well. But there's a move afoot to build these sorts of tools. And actually, in the United States, I believe it's particularly balkanized with lots of groups working on this. And so I think, especially for the higher ed community, we're gonna have to decide how do we take a sort of collective step forward to have tools that we can trust and that are interchangeable because we work across campuses and through communities of practice, right, whether that be in a professional or scientific or academic setting. So I can't emphasize enough how important this is and how important I think it is for these standards not to be developed by someone else and imposed on higher ed, but rather higher ed needs to get the table designing these in a way that meets their needs. It's a little bit of a soapbox moment, but I think there's absolutely essential for our community to consider.
Sindhu Adini: Chris and Dr. Payne, thank you so much. It was really an enlightening conversation for our audience here.
Speaker 1: Thanks for listening. If you'd like more information on how Carahsoft or Google Cloud can assist your higher education institution, please visit www.carahsoft.com/google or email us at Google-Sales@Carahsoft.com. Thanks again for listening and have a great day.