Partner Spotlight: Clinical Architecture and Verato Combine Forces for Comprehensive Data Quality

Stephanie Broderick (00:04):
Thanks for coming to the Clinical Architecture Data Quality Theater, and today we’re going to be talking about data quality around patient demographics and clinical data, and a combined solution between Clinical Architecture and Verato. A very important partner to us. And we’ll be talking about an initiative that we had where we combined our solutions to create data quality for a very large health system. So I’d like to have my wonderful panelists here introduce themselves.

Bonnie Bruner, MSN, RN-BC (00:40):
All right, great. So my name is Bonnie Bruner. I am a Client Success Manager at Clinical Architecture, and I’m the Director of our Health Systems and IDNs, and apologize for having half of a voice this morning.

Reggie Wells, MHA, MBA (01:00):
Hi, good morning. My name is Reggie Wells. I’m a Technical Sales Consultant for Verato and I reside in the Nashville area. So it’s a pleasure to meet you all.

Stephanie Broderick (01:10):
And if I didn’t say it, I’m Stephanie Broderick. I’m a VP of Strategic Initiatives for Clinical Architecture. I’m responsible for partnerships, industry focusing initiatives, academic outreach, and also focusing on business development for the IDN space. So as I said, today we’re going to talk about the combined data quality of demographic information and clinical information, looking at Clinical Architecture’s solutions for clinical normalization, as well as Verato’s solutions around patient identity and healthcare MDM. So today we’re going to have Bonnie and Reggie share information about our respective companies and the kind of clients that we work with. We’ll talk about data normalization and Clinical Architecture’s solutions for data normalization and data quality. And we’ll have Reggie talk about their patient matching and healthcare MDM solution. And then we’ll look into an actual use case utilizing a very, very large health system that Bonnie will tell you more about. And so with further ado, I’m going to turn it over to Bonnie.

Bonnie Bruner, MSN, RN-BC (02:20):
All right, thanks, Stephanie. We only have 15 minutes. I’m going to be brief, just a company overview of Clinical Architecture. We are a healthcare enterprise data quality solution. We’re focused on managing, normalizing and organizing vast amounts of disparate data. We did that both semantically and syntactically, and I’ll jump into what that actually looks like later on. You can see on the right-hand section of this slide some of the clients that we work with. So some health systems to note that you don’t see on the slide include HCA and Ascension, CommonSpirit. So we typically work with clients on a larger scale because they have more of a need, they have more disparate systems and a need to kind of bring that data together, cleanse it, and make meaningful insights out of that. If you want to learn more information about what we do as a company, you are in our booth, so feel free to come around and I’m going to speak later to a particular product that we have a demo station for. So Reggie, you want to go ahead?

Reggie Wells, MHA, MBA (03:57):
Yes, and so I’ll give you a few minutes on Verato. We were founded in 2012 and headquartered in McLean, Virginia. We are an HMDM solution. So we are the first HEM platform and we’ll speak to more of the capabilities we provide unique to that industry as well. We do service three of the top five largest health systems. So they have trusted Verato with their patient and critical data. We also have some interests from outside of healthcare. So because of the challenges of healthcare and the critical nature of that data, other industries have entrusted us to be able to bring their data, their personal information together. And then lastly here we focus on, we’re starting to see a lot of interest in key use cases there around moving to the cloud, clinical interoperability there, digital engagement, so engaging your patients as consumers, as well as provider data integrity. So we have all this data from providers, how do you keep addresses and their contact information up to date? So we focus on those specific areas and we’ll provide more details into how we do that here shortly. But first of all Bonnie’s going to provide some insights into why we’re better together.

Bonnie Bruner, MSN, RN-BC (05:20):
So there’s certainly a lot of solutions out there that have kind of a comprehensive solution when it comes to data aggregation, having meaningful insights into that data in one place. So this is about why would you choose to use two different vendors to meet those use cases? So the first two are similar. In a nutshell, we are both Clinical Architecture and Verato. We are best in class at what we do. And along with that, you’re going to get industry-leading resources and consultation regarding both of those particular pieces of functionality. And that’s a big piece of what we call our data quality journey. It’s not just the software, it’s the consultation and what we’ve seen work best with other clients and lessons learned, best practices moving forward, proven and scalable solutions. We’ve obviously worked together, implemented this together, and had a lot of success because we have already developed an integration together that is, I would say vendor agnostic, use case agnostic. There’s much more accelerated time to value, so essentially we can implement quicker. And then lastly, just having a foundational solution. As I mentioned before, that’s agnostic of downstream vendors and use cases. So both of our products sit at kind of a foundational level so that organizations can reuse that functionality for multiple different use cases, whether it be quality measures, data and analytics, health information exchange and interoperability, et cetera.

(07:30):
Alright, so the Clinical Architecture solution, we have a few different solutions. Pivot is one of them and we actually have a demo station over here if you’re more interested in that product. Our other flagship product is Symedical and these work hand in hand together. So Pivot focuses on the syntactic normalization of data. So for example, you’ll see here different kind of message formats, HL7, CCDs, even claims data. And essentially it’s brought in to Pivot. We normalize it to an internal canonical model and we’re able to consolidate and de-duplicate the data. We have some more advanced tooling to be able to validate the quality of that data and then we can output it into really whatever form of data you want. A common use case that we’re going to talk about later is bringing in HL7, various versions of HL7, v2, and normalizing that data both syntactically and semantically using Symedical and output it in whatever format you want. FHIR is a common one. We also have clients that’ll bring in HL7 and they just want to normalize their HL7 to the latest version of HL7. We also have the ability to bring in various formats and we can output it into multiple formats as well. Another common one we see a lot is OMPOP for people wanting to do research.

(09:40):
So I think that’s it on Pivot.

Reggie Wells, MHA, MBA (09:45):
Okay. Alright. So Verato again, we’re built for healthcare HMDM platform and essentially we’re able to address the person as they present to your organization in various forms. If you can advance it there we are. So we are a native cloud solution, HMDM platform. We are able to address again various perspectives of roles of a person, so persons as members, consumers, patients and so forth, comprehensive enrichment data so we can create that longitudinal record of a person. So in addition to the information that you have, we’re also able to enrich that with other social determinants of health data for the patients. And then we have best in class identify verification. So as we present to your organization, we have capabilities to be able to confirm the individual are who they say they are. And then lastly, we feed enterprise data and analytics.

(10:45):
So we’ll speak to some of the capabilities that we’re able to do that with. So in general, we’re able to connect to wherever your patient information or person information will reside, identify that information. So essentially be able to unify that identity information, enrich it, create that longitudinal record, manage the information to be able to present it to whatever system needs to consume it, and then activate that data within your workflows. And so how we do that in general, each customer has provided a cloud-hosted only private cloud-hosted version of the platform. So you have various source systems, again where person information or patient information reside with various degrees of accuracy. What we do is we bring that information together and then we link it together with something we call a LinkID. So that’s the unique belly button for each individual within your organization.

(11:50):
In addition to that, we bring the associated MRNs or other identifiers for that person into the cloud instance as well as associated demographic details. So essentially creating this longitudinal view of the patient population within your organization. So how do we do that? Again, we have patented our approach for this. We do support probabilistic and deterministic algorithms, but how we differ from other solutions in the market is based on our referential matching capability. If we look here, we have two records here. Identity record A, identity record B. If you look the names are different. Rebecca Smith, Rebecca Jones. One record has more completed demographic information. The addresses and phone numbers are different on each of these records. Any solution in the market today would think that these are two different records. What Verato has done, again with our patented approach to identity resolution is we’ve developed this curated database with every adult citizen within the United States.

(13:03):
And so if you ingest our carbon database, we now have the answer key to be able to discern that these are the same persons. So we see that at some point Rebecca Smith had undergone a name change and now she’s Rebecca Jones. We have date of birth information that further informs the matching decision as well as we have address information and phone numbers. So we have 30 years of information on this Rebecca person or patient that allows us to confidently bring these records together to say this is the same individual. Where this data comes from? We curate this data from credit information, utility, and government to essentially look for sources of information where individuals self-report, right? And so that’s where you get your more higher accuracy of information. Again, it’s 30 years of historical data, 60 million updates to this database monthly. And we have a dedicated data science team whose job is to nurture and the care of the statements. So next we’ll speak on how our joint solutions have come together to solve a real-world business problem.

Bonnie Bruner, MSN, RN-BC (14:18):
Thanks, Reggie. So I know this is a busy slide, but I think it’s important to see how this has worked. So we have a particular use case, this was a health system, a large health system that we worked with. And you can see we have all these different source systems. This says EHR, really you can think of these more so as facilities, but certainly a multitude of EHRs, Cerner, Allscripts, Meditech, Epic, all you can think of. So they sent out HL7 v2, but various versions of HL7 v2 sent it to their interface engines and there were multiple interface engines. And from there came into the Pivot platform that I just described. Pivot is also leveraging in this process Symedical, which we haven’t touched on too much, but you can think of Symedical as the semantics of normalizing the terminology.

(15:26):
So the vernacular within those messages while Pivot is focused on normalizing the syntax. And so these two together really get you a clean record. From there we call the Verato, EMPI if Verato does not already have that patient data in there, we post information into Verato. But the end result is a message outbound into, in this case it was an enterprise data warehouse. So you have a clean longitudinal record with a universal enterprise patient ID. So you can truly have that longitudinal record go in. And I think one thing that’s important to note is you can see this is vendor agnostic, and even if this says enterprise data warehouse, this could be an analytics platform, it could be a health information exchange. So this particular integration is essentially repeatable and can be accelerated for the next use case. And we do have more use cases coming down the road where we can kind of plug and play this particular architecture.

(17:00):
So just some metrics to go over this use case, I mentioned Symedical earlier and being able to automate the semantic translations of terminologies. In this particular use case, it was for HL7 ADT messages. So we were able to normalize to standard terminologies so that if this health system wants to be interoperable with that data, they can again downstream to another use case. But patient demographics, race, ethnicity, gender, all those things, allergies, procedures, diagnoses and observations like vital signs. So all that data is mapped to a standard terminology now, and we have about over a million terms that are mapped to the standard terminologies. And about 75% of that was automated through our software. For the syntax, we already kind of covered this, the v2.x to FHIR and today I think it’s actually closer to 40 million messages ingested in output today and FHIR.

(18:26):
And of course those are inclusive with the UMPI. And I can tell you we already have a couple other use cases from this client that wants to repeat this process for other downstream systems. For Verato and Reggie, feel free to jump in. Started with 4.7 million records that were historical data loaded and got those down to 3.89. So almost a million or less patients that they were able to kind of come up with. And again, across 60 sources of data, which is encompassing of really all of these results. If you know about data quality, the foundation of it and everything that comes along with it, I think you would agree a three month implementation is impressive. So essentially from installation to go live getting all the Pivot interfaces configured and all the normalization done was about three months.

Stephanie Broderick (19:43):
Bonnie, I’d like to add just a little bit here. So we did a webinar with this health system and it’s out on Becker’s Healthcare if anybody is interested. One of the things to note was when they talked about some of the results, they said because of the data normalization that they’ve been able to achieve, that they had already saved the lives of over 40 patients, which was tremendous. And that was after just a very short period of time. So can’t overemphasize how important data quality is for decision-makers in order to drive those insights into what’s going on with patients so that you can create better outcomes. All right, Reggie, before we go into questions, if anybody wanted to find out more about Verato solution, is there someplace that they can go here at HIMSS to find you?

Reggie Wells, MHA, MBA (20:33):
Yes. First, we have the Verato team here at this table, and then we have a booth in the far left corner. So just keep walking until you feel like you’re about to leave the stadium, in that place.

Stephanie Broderick (20:45):
All right, fantastic. So do we have any questions from the audience? Yes.

Audience Speaker 1 (20:52):
Thanks Stephanie. I was hoping you could talk more about the 40 lives you saved, how you know that for example?

Stephanie Broderick (21:00):
I know that because the IT director told us that from the health system.

Bonnie Bruner, MSN, RN-BC (21:09):
We have the same question from the webinar.

Stephanie Broderick (21:11):
Yes, I asked actually the same question on the webinar and they were a little bit vague, but they wouldn’t have cited it if it hadn’t been true.

Audience Speaker 1 (21:24):
(Inaudible)

Stephanie Broderick (21:29):
Yes, yes. And for us, I mean, we’re in the business of trying to improve outcomes for patients. And so when somebody uses your technology and you hear of that kind of result, that sticks, that resonates. But we try to get more details and we didn’t get that.

Bonnie Bruner, MSN, RN-BC (21:49):
I’ll just add, and it goes back to us being a very foundational level to support multiple use cases. So again, that’s population health. If you want to be proactive in keeping your patients healthy, it’s decision support, which would certainly tie to lives saved, health information exchange. The use cases theoretically should be able to feed off that initial foundation of the data quality and it makes it much more efficient to be able to get those insights.

Stephanie Broderick (22:33):
But it’s really critical when you’re looking at an analytics use case or a big data use case to be able to align your data both at a patient identity level and then to be able to aggregate at the clinical level, to manage the population. So both things are critical, and so that’s why we’re so excited about being able to bring these two solutions together into a very elegant, easy to implement solution for our clients. All right. Well, we really appreciate your attendance today, and thank you to our panelists for a great job.

Bonnie Bruner, MSN, RN-BC (23:06):
Thank you.