This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.

 Today, on this week, health it. I think that overwhelmingly care teams ensure that they're providing really high quality and low.

Ensuring that their patients get the best outcome. And what I've seen is that that makes them really primed for adoption of new tools and processes that are aimed at achieving those outcomes.

Thanks for joining us on this week in Health IT Influence. My name is Bill Russell. I'm a former CIO for a 16 hospital system and creator of this week in Health IT a channel dedicated to keeping Health IT staff current. And engaged. Special thanks to our influence show sponsors Sirius Healthcare and Health lyrics for choosing to invest in our mission to develop the next generation of health IT leaders.

If you wanna be a part of our mission, you can become a show sponsor as well. The first step. Is to send an email to partner at this week in health it.com. I wanna take a quick minute to remind everyone of our social media presence. We have a lot of stuff going on. You can follow me personally, bill j Russell on LinkedIn.

I engage almost every day in a conversation with the community around some health IT topic. You can also follow the show. At this week in health it on LinkedIn, you can follow us on Twitter, bill Russell, HIT. You can follow the show this week in in HIT on Twitter as well. Each one of those channels has different content that's coming out through it.

We don't do the same thing across all of our channels. We don't blanket posts. We're actually pretty active and trying to really take a conversation. . In a direction that's appropriate for those specific channels. We really want to engage with you guys through this. We are trying to build a more broad community, so invite your friends to follow us as well.

We want to make this a dynamic conversation between us so that we can move and advance healthcare forward. Today we're joined by Andrew Sorenson, the Chief Analytics Officer for Castell. Andrew, welcome to the show. Yeah, thanks a lot. Great to be here. I'm looking forward to this conversation. I, I'd really like to model this out.

This was one of the most challenging things we're gonna talk about value-based care contracts across a clinically integrated network. We're really specifically gonna dive into data and including all the data that is available, not only the EHR data, but. Information around the, uh, social determinants of health, payer data and all those kinds of things.

And Intermountain and Castel are making significant progress in this area. And you did a presentation at HIMSS and I, I'm looking forward to, to pulling that apart. To start us off, how did this. The value-based care contracts. Take us back to the beginning at Intermountain and, and how it started and, and how it's evolved.

Early on, I think people envisioned that it would be about a six to seven year transition period, and we set some significant milestones. I. Fortunate enough to be involved in some of the analytics early on in this process. And we saw that we could make a transition by doing things like moving from a bunch of percent of build charges, contracts to DRGs with an eye, to eventually taking on full risk.

And that transition, I think, really positioned Intermountain to accelerate the momentum that it had with respect to value-based care. And about a year later, another major milestone in this transition was the establishment of a new and innovative primary care model called Reimagine Primary Care. And I remember doing some of the analysis on Reimagined primary care.

It's interesting when you do analytics, sometimes the results are so bad that you feel awkward presenting them and, and you really go back and you pressure test them. You wanna make sure they're right. I think one of the only things that may give me more anxiety than that is presenting very, very good results because you wanna make sure that you're right.

And with Reimagined primary Care, that was essentially the situation that we were in. We saw such great performance so quickly that I was personally nervous in analytics, that the results were too good. And as those results with respect to quality and patient engagement and caregiver satisfaction started to roll in, I think Intermountain leaders saw that we needed to accelerate deployment of that model.

And so Castel was formed to help take a lot of the lessons that Intermountain has learned with respect to value-based care, and then help deploy some of the capabilities that we have to the markets that we serve. One key thing that we found was that the patients that, that we provide services to through select only about 50% of those patients receive.

Their primary care services from an Intermountain healthcare provider. And so in order to maintain the momentum that we were seeing, we needed to deploy these capabilities to to others, and so. 10 year journey that we've been on in a nutshell. Wow. So you are the chief analytics officer for Castell. Talk about the, the services.

We're, we're gonna go into some of the successes. We're gonna talk into some of the challenges, but first, lay the groundwork. What is castell, what are the services and, uh, I, I guess the analytics and, and the things they bring to the market? So Castel is both a, a product and services organization. When you think about the product side, we've got significant data and analytics offerings, and then a, a really robust set of services that we provide to care teams as well as some consultative services for groups like IDMs and others.

So thinking about those services, you can break them into a few groups. We, we have a number of services that are delivered to patients in the home. Innovative things like hospital level care in the home, or more primary care focused services, such as ho, pardon me, house calls. And then we have a significant set of services that are aimed at supporting care teams in delivering really great care.

So for example. We have a group that we call care traffic, care Traffic controllers help stitch together transitions of care. They help support care teams within the clinics that we work with to understand the patients that they should be working with, who's on their schedule over the next day, and who's not that shouldn't be.

And then a, a few other services all aimed at, at delivering better affordability and, and quality. I love that concept of care traffic control. Is that the, the next step in terms of care navigation essentially, or is that a little different? So I think that navigation is an element of, as I work with the leaders that oversee that function, what I hear time and time again is that they find really upbeat smart people that are highly engaging for patients to work with.

And so it's everything from bartenders and hairdressers, people that are just really easy to interact with and the care traffic control People occasionally are doing outbound calls to patients to check on them from a wellness perspective. They're reaching out when there are open care gaps. Let's say if there's a need for something like a breast cancer exam.

They interact with those patients and then with the clinics that Castel works with to help get those patients on the schedule in those clinics. I, I'm gonna essentially take us through this imaginary journey that I'm representing a health system that's trying to launch this, and I think we're gonna start, we're gonna start at the high level and then we're just gonna keep moving down.

Because I, I'm really curious how we pull these data sets together and how we use them. But I wanna start at really at the top, which is, you talked about this early on, that we have two different businesses running. We have a very successful hospital business, which a lot of organizations do. And then they we're looking to make this move into value-based care, which is.

A a different business and in some ways a little competitive. So if you were sitting down, let's say this is a board level conversation right now, and the board's bringing you in to say, all right, we're contemplating making this move. What are some of the challenges? What are some of the things we need to be thinking about as we move our health system in this direction?

You know, one of the key challenges that I hear that health systems are facing. It is a timing question and uh, I guess the timing of returns on the investments that a large system needs to make in order to be successful in value-based care, on the one hand, if they move into value-based care contracts, but they haven't made the investments in some of the tools and services that they need to be successful at that.

They run the risk of poor performance in those contracts, which could adversely affect them financially. On the other hand, if we make some of the really intense investments that need to be made to, to spool up all of the tools that they need to be successful, but they haven't brought the value-based care contracts online, then they run the risk of a really long runway to pay back the investments that they've had to.

And I think for executives that are in a really high performing system. Significant investments in acute care. I think that there's also a little bit of trepidation about the fact that value-based care can reduce the census in hospitals or the margin that's being made in those hospitals. So it feels a, a lot like it's a, a question of timing.

unfortunately, I think there are ways that you can ease some of the pressures related to that timing. I, I mean, Castel, for example, has built out our tools and services in a way that they'll scale really easily. And so rather than every organization having to make this on their own, we can help smooth out some of those timing issues.

We heard this as a trend coming through the pandemic because obviously the payers did fairly well and the, the providers lost a significant amount of revenue for a period of time, and they were sort of looking at how do we get more managed lives and managed contracts and those kinds of things, which.

Which requires taking risks, but it's a whole new set of skills. It's a whole new set of tools. Talk a little bit about the tools that we want in place. I would think as ACEO, one of the tools I'd want is some sort of modeling that shows how much from a financial standpoint, if we grow our value-based care, how much it, how to sort of manage that.

That tightrope between the, the two models that we have of really accentuating one another instead of pulling business away from each other. Yeah, I think that's right. You could probably think of our analytics tools in a few different buckets and, and if we set aside for a moment the tools that we're building to help people do their work every day, let's just talk about some of the tools that we've built to help executives monitor performance.

One of the. We experienced early on was that leaders wanted a lot of information in these tools. They wanted to be able to see how they were performing across their value-based care portfolio, and there would be greater demand for more and more data points over time. What we realized was that the underlying motivation for adding more and more data to these reports was that people really wanted insights and.

And so one of the things that we've done is to build a set of reporting tools that not only give people a broad view of how they're performing, but also help them very quickly focus in on the things that need their attention. And then over time we've taken those a step further to say, well, what can you actually do about it?

I think. We'll add value to someone at the most senior level of an organization because they're very busy and, and I think they want to know the three or four things that they need to be worrying about. But then they're also helpful because a tool like the one that I'm describing can help an an operator that maybe reports to that leader.

To understand what they can be doing about it, to, to capture the opportunities that exist. The surveillance process, I think, is really important. Just surveilling all of your data to look for the opportunities and using that as a mechanism by which you're describing your performance. One of the challenges we had in our program in Southern California was the, the timeliness of the data, because getting all that data together was a challenge.

Then normalizing and cleaning that data and, and then. Essentially turning it into the analytics that we needed. Uh, a lot of it was retrospective. Have we moved beyond that? Are we getting more realtime predictive? Can the, can leadership and even the clinicians look at it and get a, a more predictive model?

Yeah, I, I would call it more near realtime than real. You've called out something that we definitely struggled with early on. Definitely a challenge for a hospital operator who's used to. Very real time data can look in the morning at what happened yesterday and make adjustments and tweaks to ensure that they're performing highly.

Working with those same leaders and telling them that you've got a 90 day claims lag, which is really common. It was a new way of thinking, and so as we worked with the insurance companies that we partner with to try and get data on a more regular basis to. We've also tried to find ways that we can supplement our overall data asset so that it's more responsive and more real time.

So for example, two really important data sets that I think we, we have learned to manage pretty well with the vendors that we partner with are EMR data feeds. And data that we get from health information exchanges, and those vary a lot by state in terms of the sophistication of those HIEs. Uh, but what we've found is that if we can deliver information to people about who is in the hospital yesterday, or information about patients who need an appointment but actually don't have one, because we can see that in the MR.

It actually supplements the claims data that we're getting from payers in a really nice way. And we're able to drive intelligent enough workflows and, and reporting by bringing all of those together that it's helped close the, that gap. So data sets, you talked about insurance. Claims data. You talked about EMR data, H-I-E-I-I put into that category of part of the electronic medical record.

I assume you have other data sets, you have social determinants data. What other data sets are key for you to, to build out this program? So I think you've named really the vast majority of them claims, EMR, like you said, ADT can be connected to the EMR. That makes a lot of sense. Social determinants and data.

That we have learned over time is incredibly valuable, but less intuitively. So is information about providers in our network. And in particular, one of the things that can be a little bit complicated is getting a really good sense not only on a very up-to-date basis, what providers are in your network, but what constitutes a primary care provider because clinics and payers will think about that differently from one another.

We work with a lot of different payers and, and sometimes payers think about that differently from another, and so we've really started to invest more and more in having a. Dataset related to the providers that we partner with to ensure that we're delivering the right information to them based on the, the roles that they fulfill and that we're able to manage our different programs that are catered to value primary care providers.

So, so you guys aren't the plumbers, uh, when I call the, the data plumbers, right? So you're not, a majority of your work isn't like finding the data and moving it around you. You have partners that, that go out and find that data and bring it in and normalize it. Is that how, how you're doing it? Yeah, to stick with the plumber analogy, I think we've partnered with vendors who can build the storm drain type plumbing, the real industrial strength infrastructure type plumbing, and then we do some of the last mile plumbing.

So we've got data engineers and data architects that are thinking about how we bring fluids to and from the house, but we are not going out and necessarily building the storm drains or other plumbing infrastructure because over time the vendors in that space have gotten really sophisticated and we just didn't feel like we could do it better, faster, cheaper than, than they can.

So they're bringing the data to you. You're doing the last mile. And starting to take that data and turn it into tools and insights and information that executives, care teams, providers can use in the delivery across the value-based care contracts that you have. Yep. All right. So, so we talked a little bit about tools for executives.

Talk about the tools that you have for care teams and then providers. Sure. So care teams are, are really. A key part of this equation and going beyond the element of that which may be really self-evident, is we listened to providers in the clinics that we work with and worked really closely with the medical directors within Castel.

I think one of the common threads in those conversations was that providers are are asked to do a lot, and they're really passionate about delivering high quality care to their patients. . It's a general rule. They want the data on demand. They, they don't want it pushed to them. They don't wanna be overwhelmed with alerts.

And so I've seen my role when I put my product hat on is being a reducer and not a producer. In some ways, we want the right insights for that provider at the right time and the right setting. And so by taking that, that softer approach, providing data at the point of care in, in very deliberate amounts.

I think that we've driven really great buy-in, but the care team, like I said, is one of the critical ways that you're able to accomplish that with care traffic control or, or with other members of the care team in clinics, you can prime them with the information that they need so that they can be deliberate about what they communicate to the provider and when.

And you can also deliver some of the insights to that care team member that they think that the provider will wanna know. So for example. If we know that a patient is seeing a, a specialist that maybe the primary care provider's not aware of, we wanna make sure that they know that or if there's a new prescription.

And so we've tried to be really deliberate in priming the care team and the support structure for those providers so that they can be taking action and everyone's working at top of license. So some tools that we have to help do that. We've got a a morning huddle dashboard we call it. It answers the question of who is going to be on my schedule and who isn't going to be on my schedule that needs to be, and, and how do I take action on that?

Interesting. I mean, that, that in and of itself is, is really fascinating to me. I assume it's creating. Like a, a dashboard for that care team to work from where they can say, all right, these people go into the call center for outbound calls these people go into, but essentially it's a dashboard that they're looking at that is, I mean you call it care traffic control.

I would assume that they can look at it and see where at a higher level, where everybody sort of is at and where they're moving through the system. Is that pretty accurate? Yep, that's pretty accurate. So there's a lot of logistical data like that. And then there's data that we've run, algorithms and models behind the scenes that help them do prioritization for things like outreach.

And then a member of care traffic control can either curate or foster that morning huddle process, or some clinics prefer to execute them on their own. Again, I love that. Care Traffic control. That's, that's awesome. Have you guys copyrighted that? Is that trademarked and, yeah, we weren't able to trademark it, but we still use it and we think that it's a great way to talk about.

Of that function. Yeah. That's awesome. But let's talk about the tools for providers. Back in the day, back when I was CIO, we were working towards this concept of a whole person profile or a whole patient profile. We call this the whole person profile 'cause we wanted to move beyond patient to consumers.

And it, when I look at the data that you're bringing together. You're probably, you're probably getting pretty close to a whole person profile. A lot more aspects that, that really contribute to health beyond just the, the delivery of healthcare with the social determinants, with the claims data and whatnot.

We've always talked about the longitudinal patient record as sort of the Holy grail, but to me, the whole person profile gave the clinician a lot more information if they went into this system versus the EHR, would they? Get a better picture of, of the patient and what's going on? Yeah, I, I think so. For exactly the reasons that you've pointed out, that we're able to bring together data from 70 different EMRs, from 11 different claims, feeds and curate and, and tee that up in a way that is easily consumable, and they're able to see that full longitudinal record like you've talked about.

So how do the clinicians use these tools on a daily basis? I think it varies by care team and clinician. We, we see some clinicians that are very data hungry and like being deeply in the data. Some who are content to have their care teams take a look at it for them. But if, if a provider is doing a.

Pre-vis check-in with their ma let's say we're able to give them that full longitudinal data to say, here are the meds that your patient is on. Here's the specialist that they're seeing. Here are their last few interactions with the health system. And then we're working to think we're, we're working to integrate SDOH data into that process, although we wanna make sure that we're doing it in a way that feels both meaningful and actionable to the provider and doesn't just end up as, as one more piece of information.

I'm looking at your HIMSS presentation. It looks like you're partnered with Arcadia. Are you bringing that information back into the EMR workflow or is is there a different workflow that gets created for value-based care? So the tools that we have interact pretty closely with the EMR, although we're not at a point where we're intentionally writing information back into the EHR.

I think that there's great demand for that. I think that it'll ease some of the adoption burden, but every instance of an EMR, in addition to the fact that you've got a ton of EMRs out there, I think you said you had 75 different ones earlier. Not only do we have a similar variety, but we've also found that the different instances or installs make writing back to the EMR just a a little bit dicey.

So we're not at a point that we're doing that yet, but we wanna make it as frictionless and seamless for a provider as possible. And so, for example, Arcadia has this tool called Arcadia Desktop Itns memory packets, let's say, and understands in context that the patient that a provider is seeing and then tease up information about that patient.

That the provider can easily consume it. Earlier on, you talked about the fact that there was a lot of care being provided outside of the Intermountain Network of Provi. I mean, maybe they were part of the Intermountain Network providers, but they weren't maybe part of the EMR or whatnot. Talk a little bit about a couple things here.

One is more and more of the care is being done. Clinics in different places in the home. So how are we orchestrating all that? And then I'd like to talk to you about, are you starting to bring in a specific device data and those kinds of things as you move people into the home. So there's, yeah, there's a lot there.

So in terms of how orchestrating delivery of care in the way that you're talking about, uh, I think there are a few important things and. A few of them already, right? The products and services that help these care teams be successful. I, I think that overwhelmingly care teams want to ensure that they're providing really high quality and low cost care to their patients, and they're deeply engaged in ensuring that their patients get the best outcome.

And what I've seen is that that makes them really primed for adoption of new tools and processes that are aimed at achieving those outcomes. So it, it's been a fairly easy process in a lot of ways to find people who are engaged and, and it's more a pull than a push at a lot of times, especially as we work to.

Align the incentives that we've got for for these care teams with respect to high quality and low cost outcomes. With respect to your question about integrating additional data feeds, it's almost like you can think about a whole patient record. As a there, I think we've achieved one in terms of versions for that.

We've got the claims data, we've got the SCOH data. We've got EMR data, but I think we're moving to a place where there are these additional opportunities. There's wearables, there's remote patient monitoring data that we can be looking at, and streams of data that I think we've got the technological and analytics.

Now the question is how do you condense all of that down into a set of insights are meaningful for a care team? And so over the next few years, I think that those will be important things for us to explore. Alright. Talk to me, I, you know, I'm looking again at your presentation. You have some success stories here, uh, in terms of, uh, supporting the Covid vaccine rollout and other things.

Not only a amidst the pandemic, but also just successes over time. What kind of results have you been able to to drive through this program? So, last year was a difficult year for a lot of people, and as I think about Silver Linings from the year, one of the things that consistently comes to mind is the way in which it helped us be highly focused on what was going to be most meaningful for people.

Another example of that is the vaccine outreach that you've mentioned. We knew that we needed to reach out to a very large number of people very quickly, and we were able to partner with people not only at Intermountain, but also in our vendors' companies to find ways to deploy text messages, for example, to people.

And so thinking about the results of some of that, I mean, we sent 220,000 text messages, I think, and nearly 40,000 people who received those text messages went to the Intermountain Vaccine scheduling website. When I think about the number of phone calls that would've been required and the cost of doing that, that would've been a really expensive endeavor had we not pursued, but thinking about some of the other, and.

Versus the clinics that were not as highly engaged. We see a meaningful difference in the, the cost per patient. And, uh, that was about $2 and 50 cents PMPM in a year where all of the clinics it seemed were going down. As we've looked at similar numbers this year, we, we actually see that trend accelerating.

So we look to that as a really promising sign. One of the other measures that we've looked at is the extent to which clinics say that they feel that we're helping them perform better at value-based care. And as we see early survey results as we're interacting and engaging them on that question, nearly 80% of them are saying that Castel is helping them be more effective at value-based care.

So that's another result that I'm really proud of. I think positions us really well for a bright future. There's a cultural change, but there's also an education that needs to happen. Is that something you help with the education of, of the clinics and the clinic staff around value-based care and how to deliver it effectively?

Yeah, that that's an important thing that Castel does, and over time, I think we're seeing that, that people really understand the concepts more and more. We always start with some of the concepts in kind of your 1 0 1 level class where we all start, but now it, it's moving fairly quickly to. People wanna understand the concepts, but they really also wanna understand how can they be great at delivering care in that model?

Because again, they're really passionate about helping achieve high, high quality and low cost for their patients. What I find fascinating is you go into a pandemic year and it requires you to pivot and to be really agile as an organization. And what we found in our interviews is that the health systems that were

Progressive with regards to their data. They had really matured with regard to their data, were able to pivot and and be agile because they had the information. They were able to create predictive models around covid hotspots and outreach programs and vaccine outreach programs, and that's what you described here.

There's a level of sophistication around the data that allows you to dream up new models as they're required. It would seem to me that's one of the biggest benefits. I mean, obviously the numbers you shared are, are incredibly impressive, but just that level of sophistication because healthcare continues to.

Change. This gives you that set of tools that you can stay ahead of that curve. I think that's right. We built on a significant foundation. Not only that had been created during the time here in Castel, but really with tremendous partnership and over a long period of time by leaders at Intermountain. And so we had a great data infrastructure in place.

There were a lot of leaders that we partnered with, whether it be from a data science perspective, to supplement our own data science capabilities from a data architecture perspective. That helped us deliver a lot of these things. And sometimes the tools and applications we were using had come from great leaders at Intermountain.

So in addition to having a great foundation, it was great to have that, that collaborative effort as. So what's next for Castel?

I, I see a few things on the horizon. As I mentioned a few minutes ago, we're seeing. Accumulating results showing us that we're really on the right track and, and having high performance. I think as we share our experience and, and talk about the successes that we're having, we're seeing increases in the demand for, uh, the services that we provide and, and for our capabilities.

But then as I think about the digital experiences, I think about analytics. I see a, a big opportunity to enhance the user experience. To put together a more cohesive experience for care teams across all of the tools and services that Castell is providing. So as I look to the future, I get really excited, uh, a little bit daunted if I'm, if I'm honest, but I think that there's really great opportunities on the horizon.

Yeah, this is fantastic. I appreciate you coming on the show, sharing some of the insights from this presentation. This is one of the topics I saw at HIMSS that I really wanted to delve into further. As I shared with you before the show, this is one of those areas where as ACIO, it was, you could talk about it still being daunting today.

Our sophistication around data science, they've all increased over the last decade, and I think we're looking at a really interesting time over the next five years for things like Castel to really change the way we as patients experience, care, and experience our health system. It'll feel more cohesive, I would think.

I hope so. Andrew, thank you once again for your time. I really appreciate it. Thanks a lot for having me. What a great discussion. If you know of someone that might benefit from our channel, from these kinds of discussions, please forward them a note. Perhaps your team, your staff. I know if I were ACIO today, I would have every one of my team members listening to this show.

It's it's conference level value every week. They can subscribe on our website this week, health.com, or they can go wherever you listen to podcasts. Apple, Google. Overcast, which is what I use, uh, Spotify, Stitcher, you name it. We're out there. They can find us. Go ahead, subscribe today. Send a note to someone and have them subscribe as well.

We want to thank our channel sponsors who are investing in our mission to develop the next generation of health IT leaders. Those are VMware, Hillrom, Starbridge advisors, Aruba and McAfee. Thanks for listening. That's all for now.