Episode Transcript

Welcome everybody to another episode of compliance conversations. My name is CJ Wolf with Healthicity, and I'm really excited today because our guest, Dave Cardelle welcome, Dave. 

Great to see you again this morning, CJ. 

Thank you. I'm really excited to have Dave on the on the podcast because he brings a really unique perspective. We're gonna look at things from a payer perspective a little bit. And, Dave and I were recently at a a conference together, and I just was really excited to to see things from the payer perspective. I was kinda you know, most of my history and experiences on the provider side, and so it was great to have different perspectives in the room. And and so I was really excited when Dave agreed to to share a little bit with us. Dave, before we jump into our the topic we've chosen for today, we'd love to hear a little bit about yourself, and maybe, you know, how you've arrived where you are, what you do, whatever you wanna share in that regard. 

Great. Thanks, CJ, and and thanks for having me. I'm really excited to kinda share, you know, with your your group, you know, some of the insights on the payer side. It's something that I've been trying to do for several years now. 

I think it's super important for our industry. But first, a little bit about me. I'm a I'm a pharmacist by training, and and I've been in the payment integrity service vendor side, basically serving all the largest health plans in the country with audit services of all kinds across the spectrum or the payment integrity continuum, for probably thirty plus years. I was probably one of the first contingency based audit vendors in that space, starting, you know, way back out of Boston, with a small company, after I had served on both the provider side as a pharmacist on the payer side working for Blue Cross Blue Shield plan, and then going out and starting my own business with a physician and another pharmacist, specifically around payment integrity. 

I had helped to develop some internal payment integrity programs for the first time within the payer way back before things were called specialty pharmacy. We were going after Okay. Across drugs where there was just egregious kind of overpayments based on mostly errors, quite frankly, just unit errors. There was a lot of complexity and, you know, calculations and conversions and fee schedules and contracts as you all know very well, Within that level of complexity, even thirty years ago, there was just a lot of overpayments, and that led to, you know, home infusion, DME, renal renal dialysis. 

I mean, you name it. Just, DRG, hospital charge audits, you know, all those things back then thirty years ago that are still happening today in a little bit different forms, but, you know, have matured and been refined around, you know, the process, to get there. So, just wanted to kind of let you all know my background a little bit more than I usually do because thirty years in the payment integrity service industry. 

I've got a lot of built up domain expertise, but I really, the last four years, have really intentionally transitioned to, disrupt that. So something I helped create, I feel that we're at the at the it's commoditized. We're at the place now in the industry where it's really time to share insights on both sides and get to what's mutually beneficial to both. Because we speak this very rare language that not many people understand. 

It's so complex. So I've gone into the, software side and and, been really focused on disrupting in a transparent way content, and what health plans do to help not only providers, but health plans. So it's it's helping both. Right? 

I think there's a mutual goal that we all wanna get to payment accuracy for a lot of good reasons, and there's billions of dollars at stake here. It's a pretty big area of focus, but not a lot of people that speak that special language. So I think for the first time, you know, really having both sides talk to each other, can be super beneficial to both sides. And that's what I'm trying to promote kind of where where I am now at AMS as chief strategy officer. 

And and what we're trying to accomplish is really to kinda provide data insights that can only help both sides. 

And I that's what I really enjoyed about speaking with you at the conference recently was, you know, I completely agree. Both sides want it to be done appropriately. Right? So, you know, having worked in house with hospitals and doctors, we just wanted to make sure the claim was clean when it went out, that it was accurate. 

It truly reflected the services. And what I learned at the conference was that's what payers want. They want, accurate, you know, complete information, and and, no one wants to have to work extra on these claims. Let's just let's get it right the first time. 

So that's why I was excited to to have you on and and plus you're a wonderful person. And so that was that was great too. So, thanks for sharing that a little bit about yourself, and I hope the audience understands kinda where we're coming from. I think they do. 

And, we thought we would talk a little bit today about about DRG. And and Dave had so much information, that we could share in other areas, and maybe we'll we'll do that in the future, as well. But, let if we could, Dave, let's just start with kind of tell us a little bit the the landscape of DRG, and then we can maybe get get into the specifics of of some of the work that you're doing. 

Sure. 

Love to, CJ. DRG is probably the first go to that most health plans think about when they're thinking about payment integrity and audit because these are, smaller volume of high dollar claims, obviously. 

And and first of all, I think it's good to just table set and say thirty years of my career. You know, we're focused on the errors based on the complexity of our health care system, and the contractual complexities. And all these things that come in and all the coding guidelines and policies and rules, there's just a lot of errors that happen. 

You know, percentage wise, it's not a lot, but it adds up to a lot of meaningful dollars. So let's get the fraud word off the table. That's not where I have ever been focused. And in fact, when it comes to DRG, you know, it really is truly around just levels of experience and training and coding expertise and clinical expertise. So, you know, when we start from there, you know, there's always gonna be errors with DRG claims, period. 

Full stop. Right? And it's like, it it it shouldn't be a confrontational thing. It's like both sides should try to help each other to that accuracy, that mutual goal of payment accuracy. 

So that said, you know, I've led some of the largest teams and not what I like to call an army back in the day, an army of co of auditors. Right? And a ton of coders. So first and foremost, we felt, you you know, in the way we ran our business was this super important to have actual, you know, credentialed certified expert coders, not just, you know, a mail in, you know, certification online, which didn't even exist back when we started, but really truly dedicated coders who had a passion, for really understanding correct coding and and getting into the nuances. 

And so trying to get the cream of the crop, you know, and back in that day, it was kinda localized. So you were kinda limited. Now it's kind of opened up with remote. 

But getting the best that we could and having a very focused mindset coupled with as much technology assistance as possible. So what that really meant was sifting through, you know, just massive amounts of data that was available at the time to be able to look at claims data and try to identify where the best claims from a payer's perspective were to find an error, and and making sure that time was well spent because it's it's abrasive to providers to ask for a lot of records, especially if Exactly. You're not gonna get any savings. 

Right? 

So getting precision as much as you could, and that's really progressed over the years. But, you know, getting those DRGs in a in a place where at least you have a very high probability of finding an error that you can then document, share with the provider, and get a recovery because everything was postpaid at that time. It's now progressed where things are now moving a little bit into the prepaid, not fully, but, there is pressure on providers to turn those medical records around quicker. So we used to have a a kind of a screening process that we refine and developed over time. 

Because we were doing it for so many years, we had a lot of history. So every hospital in the country, you know, for all the big payers that we were doing this work for in aggregate, you could say, we know kind of some provider characteristics of how they build certain claims like septicemia or acute respiratory or whatever it was. We kinda knew not only what are the target DRGs that CMS posts out there, but what are the DRGs down to the level of of of of granularity of how is this one provider, you know, hospital x y z, billing this certain type of DRG because we know that they've had errors historically. 

So having a profile of what those actual errors are indicated to me and to us that, hey, this hospital does things really well on this. But these two things they're getting wrong, maybe that's just a training issue. Maybe it's that code that just doesn't understand that one issue clearly around policy. So we try to get as specific as we can. 

So we have, like, a a nap system, which is no audit potential. You know, more than fifty percent of all DRG claims or inpatient claims are already at the lowest, you know, payment for DRG. So there's no reason to even look at those. Right? 

So those are kinda, like, off the table. Then we would have what we would call map gap, which is good audit potential. Just exactly what it says. We would use, like, the common, you know, high high percentage success rates and the big dollar like septicemia and things like that as a gap type rule. 

Or let's just pick up all the the small percent, maybe five four, five percent of the ones that we just know that we wanna look at those no matter what because of high dollar and high potential for based on complexity, you know, overpayments. And so we'd always wanna look at that five percent and then the rest in the middle, you know, call it thirty five, forty percent, whatever that is. 

Also kinda moderate or MAPS. And that's where we would have clinical, you know, either nurse and or coders would look at this the claim specifically and just make a a yes, no decision on whether they wanted that record. All that said, collectively, we'd probably get somewhere and back in the earlier days, it was worse, but maybe one out of five claims would have a savings. Okay? So you're digging four dry holes, and you're getting something out of that one, you know, which are it's pretty bad, but it went on for years like that. 

No matter what we did, it was kind of, like, just a low success rate. But when you had a savings, it was pretty significant. So it would make up for that. Right? And it it's still worth doing. 

But that's a lot of what I would call unnecessary abrasion today. Right? Because today Right. With technology, AI, big data, all that stuff, we can be much more precise. And that moves up to maybe thirty three percent or forty percent success rate, you know, if you're doing it right. 

And you narrow down a little bit of what you're looking at. You get rid of the false positives, and you can also pre estimate what the savings are and make a decision like, hey. Even if I move this down to a different DRG, it's only fifty dollars, you know, or or you you can make fine decisions on whether to pursue or not. Theoretically, that should reduce the amount of audits, get a higher success rate, reduce the abrasion level and frustration, and get the savings that are necessary. 

Because, you know, after all the payers have a fiduciary responsibility to do this stuff. So, all that's I know I said a lot, CJ, but so I'll I'll kinda pause here for a second. But that's kinda like where we've been over the last thirty years, how it's gotten a little bit, but it's gotten much better, and more sophisticated in the screening, what I call screening to find those DRG claims with the higher success rates and touch throughout that a little bit on how provider abrasion plays into it. But the mindset of the payer before I pause, you know, they're really trying hard to do the right thing, but there's a lot of pressure at the payment integrity level within an organization. 

That's Right. 

On them to hit a number to grow their savings, you know, the fiduciary responsibility. 

At the senior level, the the c suite, I would tell you that most c suites, in in the big health plans and even medium health plans, they don't wanna do any of this auditing if they didn't have to. Right? But there's an obligation and a and a need to do that. So that's kind of like, you know, they want to do this in the least abrasive way, in the most efficient way. And and that's where full circle today, how do we help do this by sharing insights so that less audits could happen, but more accuracy can take place naturally. 

Yeah. That's really great kind of, level setting and background, because it kinda now brings me to my question a little bit about so what about automation? Right? And kinda itemized bill review, audit workflow, you know, and and bringing that number down, so that that efficiency goes up. What are your thoughts about kind of this digital transformation, automation, those sorts of when it comes to these itemized bill reviews? 

There's a lot in the last couple of years. AMS is a software company. That's something we do do as well as others. And so the first thing we focused on was really the the first part was the auto screening. Like, how do we really use data science, big data, and everything we know and just make it that much better using today AI, machine learning, and all of that. And so we now have, like I said, very high success rates in finding errors, you know, specifically around certain combinations of DRG, procedure codes, you know, sequences, all the things that attributes that lead to a potential, you know, error or overpayment in the coding. 

Again, we felt it was really important to have coders and that they some of the best coders helped us develop that kind of logic to get rid of noise. 

You know, there are nurses who have coding backgrounds who are great, and I've seen a few. They're kinda rare. But we've seen some firms that just use nurses to do code edit code, validation. 

We did differently. We use those nurses and clinical folks with that expertise to do what we call clinical validation of a g. So we kinda put it into two different buckets, and I think that's important as we go forward. So today, we find the highest success rate that there's a potential for some type of error based on either coding or clinical validation, and those are two separate rule engines that we would use. 

Once we get the records or really in our world, once the cost the the payer gets the record, we now have something that I've been wanting for for decades, and it's the ability to use technology to be able to auto review that five hundred page medical record or six hundred page or three hundred page. And using NLP technology, you digitize that. You run the NLP natural language processing, and within two minutes, you drag and drop that huge medical record. And in a couple of minutes, it's going to read through the lab values. 

It's gonna look through all the keywords that are associated with, septicemia or acute respiratory or whatever it is that you're looking at. It's going to to analyze whatever it is that you're looking at. It's going to to analyze that, and it's gonna determine a confidence level. This is super important. 

Right? So now it's not gonna reduce the abrasion on asking for records that come up, with no savings. But it will it what it will do is help the the reviewers do this very quickly with a confidence level. There's ninety nine percent confidence that this septicemia claim is coded correctly. 

Stop. Like, don't don't send this to an office. 

Waste the time or money. 

Don't waste the time or money. Right? So you just screen through that and you save money on the health plan side and you've avoided back and forth abrasion with the provider to pursue something that's not there, you know, and reduce appeals and whatnot. So your success rate is now in the ninety percentage points. 

You might wanna draw the line wherever you want. It's seventy percent confidence level. I might wanna look at that because it's a big claim and it's, you know, it's an it's a DRG where there's often errors. So let me look at that. 

And as they review it, the technology also surfaces all the keywords, all the lab results in a highlighted fashion in a summary form so that you don't have to go through the whole five hundred pages. You can do the review much more quickly. So I look at that and I say, you know, payers this is brand new to the payers. They they some of them haven't even heard this value prop yet. 

Right? It's very early days. And I'm thinking to myself, you know, CJ, perhaps this is something that the providers could actually use before those Right. File. 

Like, why not? It's the same technology. It better Exactly. Payer even more so because then they don't have to even ask for the records, you know? 

That's right. So to me, it's kind of like almost agnostic even though I come from a place where it's been fully biased on on favoring, you know, payers. Now I'm like, wow. This actually helps payers more than anything, and it helps providers if they want to adopt that. 

And I that's what I don't know enough about the provider RCM side of the house is do they have the ability to do something like that in a workflow that makes sense to be able to take advantage of of this information for their own benefit and, you know, kind of like a byproduct. It it helps the payer, and therefore, theoretically, that should help that relationship. 

Yeah. Such great questions. Let and, I wanna kind of explore those a little bit more, in a in a moment. We're gonna take a really quick break, and then we'll be right back to talk about some of those questions. 

Welcome back from the break, everybody. Dave's been sharing a lot of wonderful information about, kind of from the payer perspective, but we just kinda went into, well, why not do this on the provider side? Why not have if this technology exists, why not, have providers be doing this type of in doing this kind of automated review to, you know, make the claims even cleaner before they're even sent? And and I totally agree. I know that, in in my years of doing, working for a lot of, in house, providers and then also as a consultant, I think this this would be really valuable, because from a compliance perspective and a payment integrity perspective, we wanna make sure that the claims go out clean. 

Of course, providers don't wanna leave any legitimate money on the table, so they would probably want to see the technology also pick up some of those types of, those types of accounts. But, you know, you mentioned, you know, let's get fraud off the table. Yes. There are people who wake up in the morning and want to cheat, and they wanna be fraudsters. And I think both sides agree, let's get them out of here because they're bad for everybody. 

But most of the people that I work where it's more like you were saying at the beginning, it's just complex. 

And, you know, people don't have a lot of time or they might not have the know how or the resources or the workflow in their revenue cycle to kind of fit this in. But but I think if if there were some way for on the provider side for them to review this because they would say, hey. Let's also capture any money that's legitimately left on the on the table that we missed. 

What are your thoughts about that? Do you think that the the technology has the ability to do that, or is it mostly designed to pick the ones that that, oh, this is, you know, coded wrong, and it would cost the payer more? 

Yeah. That's a great question. You know, I was I mean, I go back so far that I was one of the first rack subcontractors in the CMS demonstration project. And that program specifically, we really CMS focused on unders and overs for the first time. 

And, you know, the industry before the rack was all just in favor of overpayments for the payer. That was their one shot at getting things back where pay where it was viewed that, well, providers can keep on appealing and resubmitting and did a lot of opportunities to get underpayments, you know. But the payers really had just one shot at getting overpayments. So that was the first time that it was incorporated and it became kind of and and there was, you know, less underpayments, I think, in general, but they're still there. 

Right? And it's still significant and it's still, valuable. 

And I think that the logic is very easily, you know, expanded to be able to do that, in a very balanced way. 

It's like with e n the best example is with e and m upcoding. 

Well, there's the inverse of that that's naturally right sitting there that no one's looking at. 

I can show you all the upcoding. Ten percent of most, you know, physicians are upcoding e and m, okay, to their compared to their peers. Whether whether supported or not is a different question. We can get into that at a different time. 

But there's Right. Ten percent or so that are undercoding. Right? And so, yeah, it's easy to surface that up if you have the right analytics, the right approach, and the right kind of technology to do that in scale. 

Same thing with DRG. There there's under coded DRGs. Where can those things you know? And so I think there's probably I would imagine there's revenue cycle management focus on that already somewhere. 

Yes. But but never from the viewpoint of what was built for the payers to screen for overpay over, payments. 

Maybe there's something there that we could do with our insights a little different, you know, and surface some additional opportunities, for providers. But I would just say, like, running an auto screen or to say, hey. Just early warning, you know, canary in the coal mine. These are the claims that a payer is going to ask for and request a medical record at the very least. 

They might ask for more, but this is the core of where the money is and the problems are with a high potential for them to ask for it. You know? That that's coming from the insights of a payer's logic versus an RCM's logic to try to get there. You know? 

Right. A little different. It's a little more precise because it's actually what they use. 

Yeah. Well, that's just gonna say that. It's like, wouldn't it be great to know what the payers are going to potentially request? So in other words, you send it through the screener, and it's like, yeah. You know, these the small percentage of claims, the payer, you know, ninety nine percent accuracy is probably going to ask for those records. Wouldn't it be better to save everyone some time for you to look at it internally to make sure, oh, actually, yeah, we missed this. 

Or, yeah, let's not send that claim in this way. It it this is coded wrong or, you know, working through that ahead of time. I can't imagine that a that a provider wouldn't wanna know that information. 

So it's It's a it's a self serve turnkey software that can sit right within their organization where they can log on, and their claims data would flush through the lodge, and they would flag those claims. 

So they would have immediate access to know what's potentially going to be requested. Wherever you insert that, whether it's pre bill, you know, post bill, where pre release submission, you know, wherever you wanna put that in that cycle. And I'm not I'm not as familiar with it, but I think there's a place where it could sit where they could have easy access to self serve, their own kind of review on those flag claims, early early warning system, so to speak. And I think it also as a physician, you know, there's an opportunity that if we caught something early enough, that they can actually go back to the record. 

And if there was something that should have been documented that they missed, that they didn't document, they can amend that, right, before it Right. Was out. So it might pick up stuff like that, and then it's totally legit if it really did happen but wasn't documented. You know, that's the place to catch it. 

So there's there's, like, I think, a lot of value props within here as well. 

Yeah. And just I don't know how well versed you are on the technology. I'm not a techie type of person, but it just is this is the automation looking at certain file types? Is it looking at, like could it pick up a PDF and do some sort of, you know, OCR recognition type of thing? Or do you have any insight on on that, or is that a little bit Yeah. 

So first, it's just your billing data. If you're billing, you know, however it's being billed, the bill's being put together. Right? 

Right. 

But we can use those those basic claim elements and attributes and characteristics to to run through the auto screener. 

But once the medical record is available, you know, we do have the technology to be able to NLP that PDF. It's usually a PDF format. 

Okay. 

And then the system in a couple of minutes, it's pretty amazing stuff. 

That's awesome. 

Yeah. Couple minutes, we'll review that and tell you the confidence level on whether there's, you know, supported or not, through that DRG. So it's pretty Wow. Good stuff. Yeah. 

Yeah. That that is really slick. So tell me, you know, in the little bit of time we have left, what else as far as kind of, on this automation process that we might not have talked about, that you you might wanna share? 

Sure. So well, I'll give you more of a concept and maybe it's, you know, a pipe dream. But, you know, where we're at is getting both sides to talk to each other. I think there's value. 

Sharing insights for the first time, you know, I think benefits both. So why not? Why does it have to be viewed as punitive, bludgeon somebody with audits when all you have to do is share the insight and and it'll save them a lot of money on auditors and cost. Right? 

The savings will go down, which is actually a good thing at the c suite level. Maybe not at the payment integrity management level. So the other thing I think about, CJ is maybe it's worth it for payers at the c suite level to consider subsidizing some of the cost of this, you know, technology and software. Right. 

And knowledge and content and data insights, you know, funding it or subsidizing part of it mutually with the provider to make it affordable, available, and integrated in a collaborative way where both sides have skin in the game for the for the mutual benefit. So I I'm kinda going a step beyond to say, I think there's a play for that. I think I know some executives in the health plans that might be really interested in some kind of concept like that. I think it takes a a large enough health system that wants to do that where there could be some meaningful results to wanna collaborate in this unique way. 

And to me, that would be just so exciting. Right? To get both sides willingly wanting to contribute to a collaborative effort to get things right the first time by sharing this kind of technology. 

Yeah. I agree. Do you see that working at, like, a at some sort of, like, a beta testing level, like you said, like, you find one, provider who who might be interested in in one payer, or is it multiple payers and one provider? Like, how do you see that happening? 

I think beta, you know, we we do a a minimally viable product. It's already there, but we do a beta with one willing provider who really wants to, you know, get engaged and collaborate and one payer. And I'm sure I can find the payer. 

I don't know enough about providers to do that, but I think there's so much exciting value that comes out of that proof of concept, and pilot where, that can be replicated and and and scaled in the industry, you know, promoted by both sides. 

Yeah. I think like, I'm thinking from the provider side. It's like, you find the provider who feels like the process is so abrasive right now that that the that the, you know, the the carrot really instead of a stick is, hey. You know, we're requesting you know, the payer's requesting this many records now. It's costing you guys this much time in in staff hours to pull these records. What if we could potentially reduce that by you know, we have confidence in this process showing you upfront that these are the ones we're likely to request. 

I don't know. I I could see that playing out. 

Absolutely. And I think you just have to get at the right level within the health plan to see the bigger strategy, the bigger picture, the overall gain. I mean, you know, the providers are truly the health plan's greatest asset. Right? And there's been so much abrasion created and confrontation because of all of this audit activity that that needs to change. So if both sides can collaborate on one point, if that's the most contentious right now and start to get your information to make it better, Regardless of your payment integrity goals of savings, put that aside because that's the minimized here. 

Then the real relationships and negotiations for better contracts, better reimbursement can happen. 

And and right now, you're not allowed to have those conversations because there's just too much contention, right, in a break. 

And that's what I was gonna say is, like, you know, is there just so much mistrust that it's like, who's gonna take the first step forward? You know? Right. 

But it's like feel like you take that first step forward, and and the walls will start coming down. It makes me think of, you know, the story during World War one when, you know, the soldiers were on the front, and it was Christmas Eve or Christmas day, and and they all put their guns down, and they went out and, you know, they sang together, and they enjoyed the day. And, it's like, that's what it take. I think, like, both sides would love to have the peace. 

Yeah. Yeah. 

Hey. It's Christmas is coming. The holidays are coming. 

We're at we're at an inflection point because of technology. Right? Like, none of this could happen without technology. So instead of just status quo, technology's here and improving things, why don't we think about it a little bit broader and say technology can actually, you know, drive the inflection point for things to really change in a dramatic way for the first time? Yeah. I love that. I love that.  

Well, Dave, this has been awesome. Any other last minute thoughts or or comments that or, you know, parting shots or anything that you'd like to share before we kinda close out today? 

Boy, I really enjoyed a short, short conversation, but, I've got, like, a million of these little, like, pet peeves and and and areas of kind of insights in other areas that are really key in our industry that there's more to this story, you know, like itemized bill reviews. Like, there's mandates out there to review every claim over a hundred thousand dollars or fifty thousand dollars prepay or not separately reimbursable or unbundled items. You know, we have technology now that can do that in two minutes. Right? So there's Wow. There's there's all these other things that technology can do that I'd love to share with you if, you know, if you and your your audience, wants to do more of these, I'd love to do them. This is fun. 

I do I appreciate that, and I I think we'll probably take you up on that because, you know, in preparing to talk today, you and I, you know, we threw some ideas back and forth, and you had a lot of other great, really specific ideas. We I latched on to the DRG, but you got things about medical devices and, you know, you're a trained pharmacist and, you know, transparency and e and m. And so maybe we could we can come back. I like the idea of maybe having, like, a payer perspective on the podcast, and, we could make that more of a long running type of angle on all these topics. 

But thank you. Thank you so much, Dave. Yeah. Thank you. And, and thanks to all of our listeners for listening to another episode. 

As usual, we invite you to, share with us other topics that you want to hear about. And if you're aware of speakers, we'd love to hear from our audience on who they might recommend as a speaker. So, again, thanks for every thanks everyone for listening, and until next time, take care. 

This transcript has been auto-generated. Please forgive any errors.