PegaWorld | 46:13
PegaWorld iNspire 2023: Pega AI in Claims Operations at Elevance
Going live with Pega Smart Claims started Elevance on a road to transforming claims – the center of complexity for health plan operations. Join this session to hear how Elevance has succeeded in creating a unified, guided experience for examiners as well as a consolidated platform for claims processing, using Pega AI for proactive and predictive insight-driven inventory management. Discover how the organization is delivering a digital-first enhanced experience that drives personalization and meets its promises to customers and providers.
Transcript:
- So excited to welcome you to PegaWorld. I'm Susan Taylor, I'm the industry market leader for healthcare and life sciences. And I'm here with some of my colleagues and you are going to see a build for change moment. So just a week ago we had a slightly different program planned but one of our colleagues wasn't able to join us. So as I said, we're doing a little build for change here and hope you enjoy it. We're delighted to share with you what's been going on with Pega and Elevance in the claims space. So with that, I'd like to have my colleagues introduce themselves. Ashwini.
- Hello everyone. Good morning. My name is Ashwini Gupta. I'm part of Elevance so working in senior director digital operation, and then I support the various technology initiative as part of commercial claim. I'm very excited and pleased to be here and part of this big award.
- And Janet.
- Good morning. I'm Janet Rappa. I am also thrilled to be here. I'm a senior product manager at Pega for the Smart Claims engine.
- Alright, and there are a few more people from the Elevance team we want you to meet. So here's our show.
- The end-to-end CINW Transformation Initiative is a very strategic initiative. When our executive sponsors came to us, they wanted something that will completely change our claims associate experience. On average, they have to go through 10 to 15 different screens to complete a claims process that is painful. To completely change, we wanted to take an aggressive approach in saying how can we bring all of that into one place where they don't have to context switch and they can be in one system.
- It has enhanced CIW as a unified end-to-end claim processing platform, which our 5,000 claim examiners utilize to do their work. With automated adjudication through the digital claim examiner and integration with various tools, CIW has truly bring the transformational experience to our associates.
- What this does is it eliminates associates from having to swivel out to the WGS mainframe and flipping through the different screens to get their information. We allowed all of that information into one screen.
- We're taking away a lot of their manual processing, the manual validation and CIW is going to be able to automate that and give them the answer in one click.
- We have everything that we need available right at our fingertips. We have member benefits, provider information, all the claim lines. Anything that you need is right there on the screen available for you versus having to use multiple sessions to go find that information.
- This is a tool that is user-friendly and was made specifically for our claims processors and our service people to really address some of their concerns and some of their things to make it just a a better system holistically.
- Our claims are going to be processed faster and more accurately.
- It's like lightning, seriously. I mean it's completely a game changer for claims processors and adjusters.
- So this is just the start, the information's in there, but we're going to continue to work with the teams but we're going to continue to work with the teams to do enhancements and improve the functionality going forward.
- CIW future improvements include the claims audit to improve claims accuracy, high dollar process automation, enhanced chat bot and generative AI support. I could go on for days talking about what we have planned for this platform.
- We now have a roadmap, a tool that can definitely allow our associates have a tool in their hand that is completely digital, that integrates well in the ecosystem, that brings in the latest technology and then also allows us to process claims accurately in a much fast manner.
- Very optimistic, very passionate and look forward to the wonderful opportunities that will come with the CIW application and the wonderful things that we'll accomplish together as a team.
- And that's what we live for that kind of impact. So with that, we're going to take a step back now and tell you how we got there. And I'd like to ask you Ashwini to tell us a little bit about Elevance Health and your digital strategy program.
- Absolutely Susan, I hope you all like the video. That was the real feedback from the real people. So at Elevance Health our mission is to enhance our healthcare experience, basically improving people's health wellbeing. Traditionally we have been selling health benefit plans. And now we are doing a transformation as a company and we are focusing on how we all can become the lifetime trusted health partners, right? Which is a lot. Essentially we are focused on whole health approach, right? That help us to leverage the various industry leading capabilities. And then enable those capabilities using the digital platform to ensure health, which is definitely a big transformation journey which we all are going as part of our organization. You saw this video, what you see the CIW, the Claim Intake Workflow that is being built based on the Pega product Smart Claims Engine. If you know that. Really cool feature and we have really customized that to build a very unified experience to our associate, right? Which is really mind boggling and you heard some of the feedback. This is like just one example how we bring one of the or any industry leading ability to enhance the healthcare experience.
- Great, so let's talk a little bit more maybe if you would about CIW and how Smart Claims fits into your roadmap.
- So CIW started back in 2018. Seriously it's like almost four to five years, right? Going back in the time when we all had a vision to build a unified platform where we have close to 5,000 claim examiners doing their day-to-day work on different tools. So how they all come to the one platform and do the processing, right? I'll probably take a step back and talk about a little bit some of the challenges we have, right? We as Elevance Health support many different functions in healthcare, in insurance industry. Claim is one of our core business, right? That really matters because if your claim is paid correctly, timely, your provider, your member and experience is going to be very high as well. Now to achieve that, we are talking close to 40 million claim every month, which is a lot, right? And to support the quality, some of the due diligence that we perform to ensure it's like timely process correctly all those things. We have close to 5,000 associates and they used to go through many different green screens I call it, right? Our legacy system to pull the benefit details, to pull the provider details, it's a very painful, right? So what we envision how to build a unified platform that can give a very centralized view for all the data details and then they are able to do their work end to end, right? Which is really a big deal in my mind, right? We did it as part of this end to end. That's where I'm really excited and proud of, right? With the hard work of many months or years. Obviously we have a strong vision that will probably indicate about how to take this platform from fix the basics to the next advanced label, right? And that's where we need to start looking at all this data, the user actions, right? How it is helping to simplify the claim process, the timeliness, we talked about the accuracy, which is like really another big challenge because nobody wants error, right? Whether it is like intentional, non-intentional, and the real challenge is how to avoid this in a proactive manner versus like after the fact, right? So there's a lot more that is in our roadmap that I think we can talk sometime.
- Great. And some of us remember Claim One way back when. It's been a journey, and I think you heard Alan and others this morning talk about how at Pega we're focused on employee experience so you can improve the experience of your members and customers. And that's really been I think at the core of the design of the Smart Claims capabilities, it's unique in the market. I was wondering Janet, as our product manager for Smart Claims. So credit to Janet and Ashwini here for this one. But Janet, you want to tell us a little bit about how CIW supported by Smart Claims?
- Sure, sure. So, our goal with Smart Claims is first to leverage the core capabilities of the Pega platform. We wanted to use the existing workflow automation. We wanted to leverage now this new AI and decisioning capabilities. And the goal for the claims examiners users was to completely change their experience. So goodbye to the green screens. Part of that is to present that unified modern user experience by having all of these systems have a single experience. If you are using claim system one, claim system two or claim system three, you see the same thing. This also really reduces training time and it makes that training a lot more impactful because what we saw previous was a lot of training dedicated to teaching people how to navigate around a green screen, right? How to navigate legacy technology. That's gone now. So training time is shorter and it is focused on how to resolve that claim correctly. And as part of how to resolve that claim correctly, we bring all of the data into the claim. You heard Vinca refer to context switching. That's a big focus of ours is we want that user to be able to focus on the claim in front of them. We don't want them to have to go to three different websites to figure out how to resolve it. And we also give them guidance. So there's this guided pen resolution that provides information that, that claims examiner needs to come to the right answer by guiding them to it and not offering them a choice. That's incorrect.
- [Susan] That's great. And if you think about what we did with Smart Claims is there's the ability to manage work and getting work done on manually processed claims, but the interoperability in the enterprise, Janet was talking about multiple claims engines or multiple systems in the enterprise and Smart Claims has all that capability. And when you put it together, there's a lot going on at Elevance Health with regard to claims. And Ashwini, I know you guys have big ambitions, so I'm going to let you talk to this slide because there's a lot here to tell.
- No, certainly this is a lot, right? So like I said earlier, our vision started back in 2018. We are here after four or five years, we call it like let's fix the basics, right? But with the vision of building this unified platform, which is just not giving the ability to process end-to-end claim, but how to really take that to the next level and make it like more advance by providing the right data and article, insight, leveraging some of the newer technology, the AI that we are talking across multiple industry right now, right? What it means to this platform, how do we take the right claim to process on the right time with the right skill resource, which is like really a big deal because often, when you talk about 40 million transactions or claims like every month, how to identify the one important claim that you should not miss and then that claim probably provide the accuracy and not like any penalty et cetera scenario, right? Because healthcare is a very regulated industry, right? If you do not do certain things with your timeline, it impacts dollar, it impacts a lot of other perspective. Now talking the roadmap because that's like past now looking forward, right? So we call it as today we are progressive, right? We are trying to optimize ourself as we move along, let's say 2024, et cetera. And then from there on how to become really the leading from the various dimension of like claim processing perspective, right? How to get the work done first time, right? That's where our focus is, right? I keep talking the quality or accuracy perspective, but that's important. So this year, 2023, we are focused on building the audit capability. Now what exactly it means because we are processing claim, but Ashwini, you're talking audit how it related, right? So what we do today, and this is just not within Elevance Health, but in my mind this is like industry problem. We have very limited avenues to track, okay, what is the quality of all the claim process, not manually, non systematically both, right? And to do that, we take a sample of claims and try to run some process and say it is wrong. It is like required adjustment, et cetera. Now our sample size usually it's a very fraction like 0.01% if you talk a bigger volume because it's very manual intensive, it's like a very high scale work, right? Now how we build a systematic way to do it, right? Or through the advanced technology, let's say AI, et cetera, right? And then identify those error situation upfront. So that's our one big focus that we have. We have a lot of high dollar initiatives that we are talking within Elevance. How to really have the touch points that we have on the high dollar claim, right? Because high dollar claims are really $40,000 more equal or more than that, right? And sometime it goes to multimillion dollar as well. So how we ensure all those claims are like flawless, right? Processed in a very seamless manner and in a timely manner as well, right? So that's like another focus that we have. Now outside of that, we are doing a lot of the process AI stuff, which probably Janet can speak more and it is really bringing like how the AI can come in and really look at all these claims and make this process more and more efficient going forward. So that's like the third one I can call that we are focused on. Outside of that we are lot many other smaller size initiative we call it like how to manage the escalations when you have nine different departments like internally and five more departments like externally coming to you and saying, okay, here is something went wrong, et cetera, and how to prioritize, right? So how to streamline it, it cannot be like emails or Excel file, right? So I think that's one work. And then how to enable like one platform to the claims operation team members, right? To look at all the work tracking because we have brought in all the work to one in unified platform, CIW. Now how to build the better reporting, an article perspective to give a very end-to-end view for all the claims? What it means to the prompt pay, what it means to the timeliness, right? Am I lagging in a particular market where the regulations are a bit different compared to the other market, right? Based on the claim type if you know, like national claim, local claim, the host home claim, et cetera. So there is a lot to unpack, right? So these are like some of the areas that we are focused in 2023. Now talking 2024, obviously we want to continue this audit journey, which is a bit complex and it's a huge, huge kind of enhancement not being done in industry probably as far as I know talking with Susan and other Pega experts. Like, so it's going to be really a cool feature, right? What I want to a little bit touch upon, like the event driven resolution, right? How to make the interactive events. Meaning if some of you, let's say know the claim, the claim comes in and then we do the various edit checks right? Now how that edit check is going to be translating to a particular action, right? And that action can be like system or can be like a manual person. Now how I really look at that action and make more interactive if it is manual, and then watch for that step what the manual step is taken. Is it like the recommended one or it's like the... Janet talked about the guided resolution, right? So you've got, let's say error code one and you try to resolve error code one because here is like the process for that system is telling you to do certain steps, but user is maybe more smarter and doing some other step right? So how to learn from this action and really use that data to do more and more automation, right? That's like the whole concept that we are working on or thinking through this. Obviously this is like going to come next year. So a lot more to unpack there. Talking like further down the line, right? How to really make this platform more resilient, right? So we always have from IT system perspective ups and downs, we have infrastructure problem, we have so many like parameters that may impact the system availability, right? A lot of maintenance we planned sometime there are unplanned scenarios. So how to make like seamless. Where it's again based on what you need, If it is like 24 by seven need, let's build the software based on that. If it is like 24 by five needs, let's build a software. So it's like different aspect to measure and resilient means a lot. But these are some of the steps, how to make the claims and the systems right completely to cloud will that fulfill that objective, right? So those are the concepts that we are working through. Obviously we are all in ChatGPT world so exporting the Generative AI is really going to be another opportunity that we want to pursue as, so I think that's... Again, it's a very small summary, I would say there's a lot to unpack, but these are the key ones I want to highlight.
- Well, that's a great summary. And anybody who's been in claims knows it's a magic. There's definitely black magic in claims. And for people who don't know what I describe healthcare claims as is the simultaneous reconciliation of five or more contracts. And when you think about what that could look like, if you had to do it manually, you can get some sense of the complexity of what those contracts can deliver in terms of the experience of examiners. So when we looked at trying to apply AI, because most of what we talked about now is fairly great use of the Pega platform. But now when you think about those contracts and you think about rules and you think about what is the next best action associated with a claim, when you're trying to reconcile those things, you have an opportunity to actually use some AI. So I'm going to turn over to Janet to talk about AI in the context of a claim.
- Thanks, Susan. So what we found here was that all payers are having the same problem. That problem is they don't have enough staff to pay all their claims on time. So they're hitting late payment interest, they're hitting performance guarantees, they're hitting other SLAs associated with their business. So we set out to try to find a way to solve that using AI. And what we found is that with the current methods, basically everything becomes urgent. So nothing is urgent. So how do we use AI to determine of all of these urgent things, which ones have the biggest impact and which ones should I deal with first? So we wanted to balance the demands of prompt pay, interest, performance, guarantees, and other demands that are put on claims. So claims have multiple objectives or goals. So the first thing we want to do, to do this is to define the skills that are needed to actually resolve the claim. So I have a pending claim and I know what its problems are. So now I want to look at it and say, what are the goals or the SLAs or the deadlines associated with that? And notice goals is a plural word, right? There are could be multiple goals associated with a single claim and that we needed to solve. So that's where the AI comes into this. What we determined was we needed a way to determine how long it would take to finalize that claim. And for those of you who are what I call a claim geek, what that means is not just how many minutes it takes the examiner to take the action, but to take into account all of the other things that can happen to a claim once it's pending, maybe it needs to go for medical review. Maybe there's a COB issue that has to be resolved before that claim can be accurately paid. So we wanted to look at all of those things. So the AI model predicts the number of days to finalize a claim. So once that prediction is done, the system retrieves the status of the goal. So that status might be this claim is already incurring late payment interest. This claim is part of a performance guarantee group that is close to failing it's PG, we retrieve those statuses and then we use those to calculate the pull date, right? What is the date that is most advantageous to get this claim to claims examiner? Use the number of days it's going to take to finalize so that we can pay it on in a timely manner. But we know that there are already claims there that when they come through the door, they're already incurring late payment interest, they're already part of a PG that's at risk. So we want to calculate the penalty that might be associated with that claim and use all of that data to set the urgency. And what's really important here is this is not something that a business can turn on and just let it go. We all know claims doesn't work that way. So that urgency calculation can be driven not only by the AI calculations, but by what I call business levers, right? So I have an escalated provider group, whoop, I'm going to pull that lever and increase the urgency of that. So this does not take control away from workforce managers. What it does is allow them to do their job much more effectively and efficiently. They can stop putting out fires, they can stop just trying to keep the lights on, just trying to deal with today and begin to look to the future.
- I think I'd call that an Excel retirement program myself. If you've ever been in a claim shop, that's what it looks like. But what we're talking about here is multidimensional optimization, which every single individual dimension you could maybe manage with rules on a spreadsheet, but to do it all simultaneously to understand how to add the most value for the next claim the person processes is what we're really talking about here. But it's not just the claim, right? So everybody who's been through the pandemic knows what happened to our workforces. COVID, we had the quiet resignations, we had people out ill for indeterminate period of times and 222 changes in 200 calendar days in 2020. Not that I counted. If you think about the complexity of the claims in front of our teams every day, and the complexity of managing the claim team and really trying to put those things together so that you can have everyone doing the best work to deliver the most value for the organization and for your members every day. It's a really complex problem. And it was a gray area for us to employ the Pega process AI models, but not only are we looking at a single claim, but remember there's a whole stack of 'em there and they have impact on everything. And Janet, can you tell us a little bit about that?
- Sure, so one of the things that we found was made possible by using process AI was to allow the enterprise to make data-driven decisions. There are two really powerful tools that you can use here. The first one is forecasting. So you can now forecast not only your claims volume into the future, but what skills are required to actually pay those claims. So this can help you understand what do you want to do? Do you want to try to quickly upskill some folks? Do you want to pay OT? And you're making all of those decisions based on real data that is coming from your system and you're integrated with some kind of workforce management. So you have all the data you need to make good decisions about what to do based on a forecasted bottleneck. The second one, and the one that I really like is this running simulations. So in a perfect world, you just pay the claims as their priority goes and you just move on. But in the real world, sometimes a provider's going to insist that you escalate their claims. So how do you understand what impact that has, right? So this is where we get the simulations. So you can run what I call what if scenarios. So what if I took all of this particular provider's claims and I increased the urgency so that they will get paid more quickly? How does that impact the rest of the claims? What is the financial impact? What is the impact on my workforce? Do I have to pay overtime? All of those things you can now use the data that you get from the model to make those decisions to make them well.
- And again, it's all about high value, high volume work optimization. And when we first started talking with the Elevance Health team about it, one of the things they realized is this could help them give more people Christmas week off. And that's a huge win, I think in the workforce optimization. So I always ask one question at the end, which is, what did I forget to ask? So Ashwini, what did I forget to ask you about?
- I think one thing you forgot to ask about, how did we really achieve all these great success with this end-to-end work? And what I want to add to that question in my mind, right? The journey was not very smooth. We always have the bumpy rights rollercoaster, et cetera, but I think Pega, Elevance Health working together, the right set of experts coming on the table in a timely manner, building the right solution, right? So it fulfills the objective that was really, really great. Now it's just not about, okay, you build a software and then you hand it over to the business partners and use it, but how to really help them to get used to with this newer software, which is like somebody working in two decades or more in green screen and you're giving another shiny object and say, "Hey, you stop going to the green screen, come to this, right?" It's not going to be a one day transformation, right? So we have gone through a lot of such journey experience as well where we provided them the right training, the nesting period I call it, and then how to make it more successful because end of the day we are building a software and that software is used by X number of associates. If X is like 100%, we call it success, right? So I think I just want to mention that.
- Great, and Janet, you've been part of the project from the very beginning. So what would you share?
- One thing that was really impressive to me at Elevance was their transformation to an agile methodology. So when I first started working with Elevance Health, they were transforming into an agile shop, and I have been blown away by how successful that has been. I see them as a truly agile shop, and I think that's really challenging with a really large organization.
- And we always say in the claims team that we only hire customers we have fun with. So my thing is we have had a lot of fun on this journeys, it's really impactful to the team. I think we said lightning, someone said lightning. It's been really great, a really great journey and this is our build for change session. So I want to really thank Ashwini and Janet for the great presentation they've given and let's give 'em a round of applause and ask some questions. All right, so there are mics here in the aisles if you'd like to ask some questions, please feel free to step up.
- Good morning. Thank you for sharing your experience. I'm curious about, I've seen the green screens before in another application, so I'm familiar with those, but the desk levels that an examiner would typically leverage to resolve issues with the claim, how did you integrate that into the Smart Claims experience? And then also curious how you're managing the upgrades with both the mainframe and Smart Claims, that journey, if you will.
- Why don't I take the first half and you take the second half?
- Yeah, for sure.
- So, our guided pen or guided event resolution almost negates the need for job aids because the instructions or the questions or the path that that claims examiner needs to follow is embedded into the actual event or edit.
- But I do think when there is a need for a reference document for which I assume that's what you're asking about, Tom.
- Yep.
- Those are immediately accessible and they can be keyed to the event that you're trying to process, right?
- [Tom] Yep. The medical records and all that, it's pulling it all together with, yeah.
- Yeah, and pulling in data, like medical records, the great thing about the UI and the data model here is that it's all extensible. And that's part of what was done actually right from the very beginning. I think Ashwini described it as customized, we would say extended, but it's same thing.
- So I think just to add to that response, so I talked about at Elevance Health we have some of the green screen systems, right? That does like, let's say the education, to your point. But when it comes to the claim examiner, claim examiners is not doing adjudication, they're processing, they're making sure the right data elements are tagged, right? The right reference documents are looked at, right? So to do the right pricing, et cetera. Now, all these features we have built on this Pega platform like CIW. What it means, to your other part of the question as well, when you do a change workflow change do change your backend system and the frontend system. So I think the short answer is for time being yes, but eventually we are shifting into a more and more with the event code concept, let's say, right? And how to make like Pega doing the full end to end, not the backend process as well. So that's like more to come.
- [Tom] Thank you.
- So I have two questions. One for you and one for you, Janet. Ashwini, the first question is the claims, is it just a healthcare or other industries that we can extend? That's my first question.
- Smart Claims Engine is built specifically for healthcare. It can be used to to handle workflow for medical and dental claims. And we've also exploring some use cases with revenue cycle management. So on the flip side, on the provider side.
- But what we're talking about here, sorry to interrupt. The concept of the multidimensional optimization of work exists in every industry, right?
- Yep.
- This is a great example because it's one that's incredibly painful in healthcare and if you're a patient with a serious medical condition can be really impactful in your life, right? So it was a great place to start. We had a great partnership. But to your question about can it be used in any industry, the underlying process, AI concept certainly can be, and Anthony Leonardi, right there is the process AI guy.
- Okay.
- And he's here to follow up with if you want.
- Sure, then my second question is like you're talking about 40 million cases or claims every month, right? So not all the claims are same, right? Which features of Pega, how Pega helped you to identify anomalies some of them are out of way versus some of them we can just process as ease and how are you doing that?
- So there is always a... I mean you're absolutely true. 40 million claims all are identical. I can process through system no folks required, right? Like manually touch. So I think that's not the ideal situation. Like we have, we call it let's say the multiple types of claims. It could be, let's say the home host scenario, right? One classic example and where there's a person who is having the insurance in Nevada, but that person traveled to Florida, right? And how to deal with those kind of claims, right? So there's a different workflow altogether, whereas somebody having the local claim, okay, I'm living in Nevada, I got some treatment or service with the provider in Nevada that's processing is different, right? Because there are a lot of, like I said, it's regulatory process also sometime, right? So we do a lot of focus on that, I call it quality. So I think when it comes to the Smart Claim Engine to your question, so it's not about like, okay, Pega is doing something different that is in build capability in Smart Claim Engine, right? It's like how to configure those different types of claims so it can go to the different skilled resources, right? We call it let's say work basket, work queue, et cetera, right? And then all this work basket or queues are assigned to the different set of skilled folks, right? So that's the configuration that we do in Pega. So going back to Janet's point, the right claim is going to the right person at the right time.
- Thank you.
- Okay. Thank you.
- Go ahead.
- Alright, good afternoon. Just two questions. What's the percentage of claims that are processed first passed without manual intervention from your perspective?
- At this moment I would say we are close to 91.3 approximately.
- Okay, second question. Are you guys using any robotic tools outside or inside of Pega to perform auto adjudication of edits? Or does Pega have capabilities to handle the fallout edits excluding folks that would manual intervene with processing the claim?
- Yeah, no, I think that's a great question. So what I can tell you, right, we have a series of automation through a UiPath like the RPA we call it, right? The set of macros. We have various sort of the bot, but within Smart Claim Engine, right? Like the Pega offers a lot of the bot capabilities as well. So what we are doing as part of one of the initiative, okay, let's bring all these automations and then put it under CIW bot process. So it's like becomes a centralized process, right? Every and anybody can access, we have a better tracking, right? So I think that's our future state, but at this moment we still have, let's say the three dimensions that we are leveraging to automate things.
- Yeah, okay, thank you.
- Yep, you're welcome.
- Two part question. So in your 2025 strategic vision slide, you mentioned revamping the escalation process. I was just wondering what that looked like and also how you continue to calibrate that based on changes in your client landscape.
- So that's part of 2023 and it's one of the critical aspect because the way we are going to tackle this, first let's make a one stop shop, right? So you have the multiple channels, there's a legal team, there is a compliance team that is external regulatory agencies who it's like keep dumping and saying, hey, here is the set of claims and you need to prioritize this or fix it if there are any error or do analysis, et cetera. But there are all these players and to Janet's point, if everything is priority, there is no priority, right? So how to make let's say the one stop shop with the better UI. So everything comes together, have the super SMEs I call it, right? Looking at all these tickets and saying, okay, this require the priority one A, not like one but one B kind of thing, right? And then the one A goes to the right person to take an action. Right now what happens, we usually try to deal, let's say with emails, excel files, like a lot of other manual means, which is like not very effective and somebody ask, "Hey, where's the status of this?" "Oh my god, I forgot to check my emails, I was in PTO. So I think those things can be eliminated, right? So I think that's where we are enhancing Pega through building some new UI and building a backend repository to store all those things so everybody can see the status, the submitter, the person who is working on right. All the partner areas as well.
- Okay, thank you.
- Yep, welcome.
- Alright, I think we have time for one last question.
- Just one question. First of all, I think congratulations, this is a great achievement, but I was always wondering in the claims area, the tool is only as strong as you get the data, right? So the incoming data, the incoming claim submissions, you might be having all sorts of different channels like the web, email, mobile, I don't know. But have you run into any issues where the data collection is a challenge when the claims are being submitted? Once you get, I think yes, the Pega tool is smart enough to apply the AI logic, the model is great that, but you need to feed that data. Have you had any challenges around that data consumption?
- No, I think that's, that's a great question and I think somebody asked that, what is your first pass rate? And I answer that number. Now what that number indicates and include the unclean data as well, right? We are in the process of how to not have the bad data coming in the system because if the bad data comes in system, your efficiency of processing goes down, which is not the right thing to do, right? So a provider mistype, let's say their address, right, it happens, let's say somebody living in Nevada and going to California, it's a bounty condition, a lot minister situation, right? So how to really filter those elements we call it like say HIPAA level seven validation upfront. So stop those claims upfront at the door, right? We call it, it's a claim intake process and we are working on that and then send it back the communication to provider or members and say, hey, you got to fix this, this, this elements and then resubmit it. It saves their time, right? It improves the turnaround time overall if you think right. And more satisfaction as well. So definitely that's like a work effort that we are focused right now, but there's going to be some runway to accomplish that.
- Right, yeah.
- No, we do end up in those kind of situations where we get paper documents and we need to process those uncleaned data. Did you explore any Pega OCR capabilities for processing documents?
- Not OCR. We do have some other OCR technology that we use, right? Let's say there's a transplant claim, it comes with like thousand page documents. To explain all the nitty gritty. Now how to read all the data elements and tie back because what happens claims comes in day one and the document comes, let's say the day two, right? And they do not relate always. And there are so many such examples. So how to read all those data elements. So there are some other advanced OCR technology that we are evaluating as well, right? How to scan every single document and make more digitized data because more digital is better for everybody in my mind, right? And it's not like a mandate, but I think we believe if a provider goes more digital, the pricing going to be more faster and more accurate as well.
- Thanks.
- Thank you.
- Alright, we have time for one last comment. I know people are trying to get to the next session. But please go, Roy.
- Remind me of the last question.
- Sure. So I see CIW is processing 40 million claims in a month. So what is the processing time in an hour? How many claims you're processing?
- I didn't catch that, I'm sorry.
- Peak rate of processing for an hour.
- So I would say we do have, again, a lot of backend processes, right? So if I convert that data into daily volume, it comes close to let's say 1.6 to 1.7 million claims. Out of that, almost like I said, 91.3. So roughly, let's say 92% goes through a systematic process that's like every day work minimum, right? So we are talking close to 1.5 million claims getting finalized on a daily basis minimum.
- Sorry to interrupt. Are you asking the context of trying to get to real time?
- Yeah real time. Pega is handling five million claims in a single day. Is there any spill over day or something like that?
- So in our lab we do real time testing. The engine was designed for real time.
- Okay. And we're under 300 milliseconds for a claim.
- Oh, 300 milliseconds, okay.
- Yeah, and that's at a sustained rate. So we have some lab data that we can share with you in a follow up if you like.
- Okay, sure.
- But it's truly real time. The thing that always gets in the way is stuff goes wrong in the claim and then it depends and all bets are off. All right.
- Thank you. Yeah, thank you all for staying and coming and thanks to our speakers.
- Thank you.
- [Ashwini] Thank you so much.
Related Resource
Produit
Une révolution dans la conception d’applicationsOptimisez et accélérez la conception de workflows, grâce à la puissance de Pega GenAI Blueprint™. Présentez votre vision pour générer votre workflow instantanément.