Melanie Izzo: Hello, and welcome to Enlyte’s expert interview series, your go-to source for smart strategic insights at the intersection of data and claims management. Throughout the year Enlyte experts Rebecca Morgan and Jim Harris will dive into how data can be leveraged to spot risks early, uncover emerging claim trends and drive actionable decisions that lead to better outcomes. Our mission is to equip industry leaders like you with the tools and strategies you need to tackle your toughest challenges in workers’ comp and auto claims head-on.

In this first session, Jim and Rebecca discuss how today's leaders can harness the power of data to proactively prevent delayed recoveries and accelerate successful claims outcomes. Let's dive in. 

Rebecca Morgan (RM): Hi, Jim. Welcome to the podcast. 

Jim Harris (JH): Hey, Rebecca. Thank you. It's always great to be here with you. 

RM: So today we're kicking off a really important conversation. How leaders can use data to prevent delayed recoveries and accelerate better claim outcomes. 

JH: Absolutely. And as you know, data is one of my favorite topics. 

RM: Yes, it is, and mine too. I think we both have a lot to say about this one. 

JH: Most definitely. And you know, this actually is a very timely one. Just this week I was talking with a customer who was wanting to take a deeper dive into how we're using predictive analytics to identify potential high-dollar, high-risk claims. 

RM: Really, really important. Well, let's start with a question. So, question for you. How are we leveraging data and predictive analytics to identify high-risk claims early in the process? And then as the claim matures? 

JH: Sure. Here at Enlyte, we've actually been using predictive models for decades. It's so important to get on a case for case management early on and identify those potential high-risk cases. So using predictive models, artificial intelligence, we were actually using that stuff even before it was cool and trendy like it is today, but we've had models in place where we're taking information and data from our customers at the earliest onset of the claim and using these predictive models to predict whether a case is appropriate for case management or not appropriate for case management at the time.

And then the other important piece I think is being able to identify as those claims start going south in what we might call mid-case triggers. So, using data from bill review, from pharmacy, from utilization review, really all the different programs that we offer, they can provide some insight into identifying when claims are, like I said, going south and need some additional attention. So, kind of collectively using all that data together in predictive models, is really helping us to get on these cases early. 

RM: It's all about early intervention and getting that full picture of the claim. So, once we've got that picture, how do we communicate this information to the clinical staff so that they can turn the data into action? 

JH: Yeah, absolutely. I think it's important that whether that's our clinical staff or adjusting staff that you have tools in place that allow them to be notified of these changes and things that are going on. So we have a tool called Engage that our case managers work out of where all of the alerts are coming into one place for them to manage. 

So, whether that again is an early intervention, new claim type alert, case moves from medical only to lost time. Maybe surgery is getting involved. Opioid medication things like that. Having all that data in one place that the team can work off of, really helps to consolidate that and keep their focus on what needs to be done, which is getting on these cases. 

RM: So, you know, we live in an era where we just have an overabundance of data. How do we mine for that data and then share it with the clinical staff? 

JH: Yeah. You know, bill review is also one of my favorites. It's so rich with information. 

RM: Mine too. 

JH: It's so rich with information in there, and so we mine that a tremendous amount both proactively like I said, with our kind of mid-case triggers, but also retroactively as well and kind of looking at what are the trends? What kind of injuries are happening? Where are your costs going? Are there certain states, jurisdictions, treatment types that are, you know, causing your costs to go up? 

So being able to do that pharmacy again is another great real-time data piece where injured employees are filling prescriptions and having that ability to jump on those cases right away. If an opioid prescription medication is ordered, it allows us to make sure that it doesn't ultimately get out of control.

So, both prospectively and retroactively taking advantage of this data is so very important in keeping your arms around the programs that you have and what's driving your costs. 

RM: Very, very important. 

JH: So, Rebecca to get into this further, let's talk a little bit about what future data initiatives our organization is planning to further improve injured worker recovery. 

RM: Yeah, definitely. Well, you know, Jim, you and I, we both grew up professionally in the era of structured data and relational databases and SQL. And you know, you kind of get what you get there, right? Whatever data you know, you could get out of systems that's all you had to work with. And, you know, for years we struggled with the problem of, well, if we could only get more data, we could look at this kind of data point. Or we could, you know, surface risks better. Or we could identify claims that were going off track if we had more data. 

Well, that is no longer the problem. The new frontier here for us is unstructured data. Now we have so much data that the challenge is wrangling that data into something meaningful. And you talked a little bit earlier about, you know, medical bills and the richness of the data associated with medical bills. Well, part of that is the medical records and how can we use them to point claims professionals to the right decisions. 

Many times, there's hundreds or even thousands of pages we've seen of these records, and it's just too much for any single human being to consume. So how do we draw out the relevant points that otherwise might be missed? How can we use them to suggest the right care for injured workers? 

So all of this becomes kind of a combination of the structured data that we've still got that we've always been looking at and then you know, we accumulate all the best practices and best outcomes with the unstructured data to help us suggest the best path forward for injured workers. You know, and how can we use them to automate very manual processes today with the hundreds of pages that we've got, you know, start to pull out those relevant details. Surface them up so that the claims professional has the right insights at the right time. 

JH: Yeah, you know unstructured data. I'm so excited about the potential in the future of this. Medical records, claims adjuster notes, I mean, there's a lot of information that right now everybody is having to manually kind of review and the opportunities to take advantage of that is really, really exciting. I think also you know when I think about future initiatives, I think we can't really sit on our laurels. We have to continue to advance even the existing things we have.

So, looking at our existing predictive models and using machine learning to constantly let them learn and grow and improve with the more data in, the better we're going to be able to have that opportunity to get more accurate recommendations coming out of there. So, I think there's, you know, a lot of exciting information that's on the horizon for us. 

RM: It's crazy how good the technology has gotten just in the past few years, there's been an acceleration of this tech that has allowed us to do so much more with the data we've got. 

JH: Yeah, speaking of that we do have a tremendous amount of data, you know, here at Enlyte we have offerings, so many different programs, and so many different ways to help manage claims and control costs. So, I want to talk a little bit about how we are integrating data from various sources, whether it’s from the customer, from various products, to really create a comprehensive view of the injured worker and their recovery journey. 

RM: This is where integration becomes really critical and it's not a sexy topic, you know. We can't flash pretty screens up there as we talk about integrations, but it is super important that we perform integrations properly. And that all of the data comes together because it provides a full picture of the claim. It provides accurate risk alerting. You know you talked about risk alerting earlier, better, more informed decision making, better cost containment, better treatment and ultimately better outcomes on the claim when you can see the full picture and have it from end to end. Everybody that's touching a claim.

We've got a lot of great integration tools at our disposal from APIs to data warehouses to AI. I'll give one example of where this comes together visually. We have a treatment calendar in claims examiner portal where we map all of the medical treatment that's happening on a claim in a visual format. It's on a calendar so that you can very quickly see the types of treatments that are occurring and where they're occurring on the timeline. You can see gaps in treatments and you can see patterns in treatments and anomalies. It’s really a useful tool to take what is really a tremendous amount of data that would be very difficult to surface insights from if you're just looking at a grid of information. But puts it in a visual form, that we can consume, and that's what it means to bring that data together on the back end first and then on the front end so that a user can consume that data. 

JH: Yeah. And, you know, we're actually taking that a little step further. Because I do love that treatment calendar idea. And one of the things that we're working on right now is creating a dashboard and a view that is combining not only the medical treatment calendars, but also all of the additional services that may be provided to the injured workers, so integrating not just medical treatment, but case management information. You know, when was the first date of disability? When are they expected to go back to work? When was utilization review done? Was surgery done? Was opioid medication done in reviews from the PBM? 

So, really taking all these different products and all these different services and laying them out in an easy view that adjusters and customers can look at to really see what's going on with these cases. And these dashboards also give you the opportunity to bring to the surface kind of the outlier claims, the ones you know that are really old, really expensive, things along those lines too. 

RM: Yeah, that's really important for you to give that kind of full picture, end to end, from claim inception injury, you know, clear to claim closure what's happening here. 

JH: Yeah, I'm really excited about it and I think you know, data visualization is so important as well. And I want to spend a few minutes with you talking about how we're using data visualization to help the adjusters and case managers and others make informed decisions. 

RM: Like I said, just a few minutes ago, our problem today is not lack of data. It's overabundance of data. It's how to wrangle the data we've got, and it will never do to just present volumes of data to an end user. We just have too much data for that. The data has to be coalesced into something meaningful. Providing smart guidance to adjusters. Surfacing the salient points so that they don't have to dig through the volumes of data and inevitably will miss important points that are in there. Miss the patterns that otherwise they might not be able to detect.

We'll automate what can be automated, but the key is to guide users on the rest. Anything that really is going to benefit from that human touch, we want to guide users on that. And coalescing that data into something that's visual and understandable. We talked about the treatment calendar, the claim counter. Those are great examples of how we can guide users to understand the data. 

I'll give another couple of examples here. In claims examiner portal, we have what's called the blue man. It's really just an outline of a human body and it happens to be blue on the screen. So, we like to call it the blue man and we highlight on that human body the parts of the body that have been injured. So, then a adjuster or a nurse could very easily understand what's going on with the claim. Especially if you talk about an adjuster, who whose background typically is not medical. Understanding the severity of those injuries and all the medical terminology that's embedded in a diagnosis code description can be tough to understand. 

So, we categorize them, and we show them on the body and then it starts to makes sense to them. What's going on and they can start to make better decisions. And then if we take it a step further, we give them what we refer to as findings. This is just trying to surface insights about the claim that we're seeing. There was a significant gap in treatment, or treatment that started before the date of injury. Those would be important things to pull out and if you have to just dig through the data to find them, you might miss them. 

The last point that I'll make here on this is around back to the medical records. Being able to summarize those medical records. These prior examples that I've talked about are more about that structured data. Now we talk about unstructured data and how we look at those medical records and using generative AI, right, we're able to now highlight salient points for the user. And they may be different points for the adjuster than for the nurse, etc. But we can link that back to the medical record, really help point them in the right direction to answer the questions that they need to, make the best decisions on the claim, leading to the best claims outcomes. 

JH: There’s that old adage that a picture is worth 1,000 words. And I think when it comes to data visualization it’s so true. 

RM: Amen. 

JH: And you know, when you're a data analyst or a business analyst and you're diving into the details you're really able to see the story come together. But when you're presenting to an audience, they have not had that insight into looking deep into the details. So it's really key whether it's a presentation, whether it's a dashboard tool, to be able to visually tell that story. And everybody's different too, some people are very data oriented, some people their eyes glass over when you start talking too much about that. So having tools and visually showing trends and information in an easy to consume way is so, so important, I think, in the data visualization aspect. 

RM: It's how we consume data today. It's just critical. Well, Jim, this has been such a great discussion. I always appreciate your perspective and your passion for using data to drive better outcomes. 

JH: Likewise, Rebecca, I think these conversations are so, so important. The more we can help leaders see the power of early insights, I think the better recovery journey it is for everyone involved. The injured worker, the TPA, the carriers, those at financial risk and those at personal risk as well. 

RM: Exactly. Well, and for our listeners, if there's one take away from today, it's that predictive analytics isn't just a buzzword, it's a tool that when we use it thoughtfully, can really shift the trajectory of a claim. 

JH: I could not agree more, Rebecca. It's all about being proactive and not reactive. 

RM: Thanks again for joining me, Jim. Always a pleasure. 

JH: Anytime, I look forward to the next one. 

RM: And thank you to everyone tuning in. We hope you found today's episode helpful and maybe even sparked some ideas for your own teams. Until next time. 

Unleash Your Path to Success

Success is only a click away—visit our curated resource hub where insights and strategies await to conquer your toughest claims challenges.

Claim Your Future