Smarter Revenue Cycle Analysis Starts with the Right KPIs

Awesome. Thanks so much, and thanks everyone for for attending. Coming back to just the the title slide here for just a second, you know, smarter revenue cycle analysis starts with the right KPIs. Really excited to be chatting through this topic.

One, it's super near and dear to my heart, kind of as as I've seen revenue cycle in three different perspectives. One on the the provider and operation side of things, then from a consulting standpoint, and then the third from a technology standpoint. There's a lot of analysis. Claims data is super powerful.

So we're gonna walk through general objectives for today is we're hoping that by the end of the session, you'll have a better idea of of how financial reporting and targeting KPI reports can provide evidence and influence some of your strategic initiatives and ultimately improve your not only your revenue performance, profitability, margin, kinda everything that's downstream throughout your RCM. And we'll split it up into four key sections. So the first one, little bit of industry overview of how you can get to examining revenue leakage, what systems are in place, what things should you look out for, what are some of the gotchas throughout kind of your end to end life cycle that might lead to revenue leakage?

Then add some quantification behind that. Right? How do you actually from a you know, everything in health care is so focused on, there has to be a cost benefit. There has to be an ROI to things that we do because you could spend your time doing a thousand different things.

It has to be ROI driven. So what is the impact around prioritizing some of these KPI reporting and revenue cycle analysis? And then from an analyzing initiating change, like, where do you begin to start with that? There's a lot of penetration points in the end to end RCM that we could begin to tackle.

So how do you prioritize those? And last but not least, we like to to wrap things up with actionable takeaways. Right? As a as a full recap on those three segments, what are the three things that or what are a handful of things that we should be looking at to immediately get started?

So with that, like, let's go ahead and dive into that first section of examining revenue leakage. So we think of, In every PM system, every, you know, every cloud on prem, every EHR, we all have some form of analytics. And then you look at this cycle. Everyone has business intelligence.

Right? It's a it's a very vague term, but it's basically used to capture reporting. Who does analysis? How do you do analysis?

And a lot of systems and a lot of provider groups that that we've seen, they let in maybe they it's a third party analysis tool. Right? You've probably heard of, like, Tableau, Domo, BI tool, Power BI. Like, a lot of these are just reporting layers that fit on top of if you have a data warehouse or a data lake, allows you to pull and query certain data.

But these systems aren't necessarily prebuilt for the complexities that are health care. Right? And this I like this little word diagram because it's the overwhelming acronyms and mnemonics that we deal with every single day that you know, throughout health care and revenue cycle.

And there's so many specific data layers in healthcare that, especially when you look at the quantity of fields that come from the claims that are on the remits that the payer provide, and we look at all that data that can be not only does it get used in some of the analysis, but it is so critical to getting analysis, you know, quote, unquote, right.

And if you only looked at charges, payments, adjustments, that's really just a a sliver of all the data you can be looking at. Right? We have CPT and HCPCS codes. We have ICD diagnosis codes. If you're at the the facility side, a lot of your stuff is gonna be on the DRG side of things. Then you have all those ancillary data elements, like diagnosis code pointer. Like, That's a funny thing that exists on claims, But it could be a sneaky reason that claims are getting denied, and your leading revenue is allocating diagnosis codes to procedure codes.

So there's this layer of of complexity above your standard charges, payments, adjustments, allowables that exist in health care, and it's the data model in and of itself. And one thing that we've seen kinda recently, Has anyone heard or been following these payers' new policies around, downcoding and recoding. Right? Rhetorical question, but if you if you have or you've already started appealing some of these or going through, I'd love to hear that in the chat and q and a, is some of the ways that claims analysis in health care might get a little sneaky is you're beginning to evaluate two separate transactions.

On the front, you have what's called an eight thirty seven claim. Right? That is what gets generated and says, as a provider and a coding team, this is what we are billing for. Then on the remit, the payer, in most cases, should be saying, We've received your claim and your request for payment for these CPT HCPCS codes, we are paying this, or we are at least denying this.

There's been this revenue slippage where one of these common ones is you're billing a like, a nine nine two one four or nine nine two one five or nine nine two zero four, and the payer, especially Aetna and Cigna were the first two. Now we've heard that United and Humana have also hopped on this bandwagon.

They are not really telling you that they're doing anything. They're just coming in and saying, you billed us level four. We're gonna adjudicate and pay a level three.

And a lot of PM systems are just flagging this or not flagging it because they see a c o forty five contractual adjustment, and they think all is well. But in reality, there's usually about a twenty to thirty percent reimbursement difference between a five and a four or a four and a three E and M code. That you're now leaving twenty to thirty percent of the revenue for that encounter on the table. So this is just an example of the certain nuance that we've seen in health care onto, like, you're comparing two sides of that transaction between the claim that's going out and the remit that's coming back, and you're connecting all these to give you a a holistic view of what's being paid, what's not being paid.

So so I I have a couple murmurings on this already. So sounds like this is not terribly new to many of you that payers are gonna gonna begin to do this. And given the recency of, you can see just sneaking behind here, the Cigna policy is just a screenshot from their from their policy. It was effective ten one.

So as you're looking at, you know, what are we, about two thirds of the way through the month, you're gonna start seeing these remittances come through. And in q four of this year, you might see more Cigna down codes, something to be aware of is comparing what you're billing to what they're adjudicating.

So a little policy that's you know, health care is hard, and reporting and data on health care can also be complex.

And then The existential crisis is denials continue to be on the rise.

Right? So the kind of that lighter pink color. That's a survey from a state of claims report done in twenty twenty four comparing it back to twenty twenty two. And we know that denials have always been, you know, in the last fifteen plus years have been problem They have been growing. They've been getting harder to appeal, harder to resolve, and harder to prevent in the first place.

In some cases, payers actually aren't even sending back a denial. I was working with a group that came back and said, well, this specific payer for medical records, they're not sending it back as a right. It used to be a c o sixteen claim locks info and a remark code requesting medical records. They're saying, well, the payer is not even sending back an eight thirty five remit denial anymore. They're sending back a letter that's correspondence and saying, We're requesting medical records on this encounter for this patient. And their revenue cycle in most of our RCMs are not built in a way to handle correspondence as part of the workflow.

It's structured more towards when an eight thirty five denial comes back. Sure. Then it flags and says, oh, a denial transaction happened. We need to appeal this or send documentation.

It's kinda slipping through the the cracks of who's handling correspondence? Where does this mail go? Where does this digital correspondence go, and how does that get put into a a work queue that the team's able to identify and resolve?

So we know, like, second bullet down there, claim denials, sentiment, forty two percent of respondents say, in twenty twenty two, denials are are increasing.

Right? So a little less than half versus almost three quarters in twenty twenty four feel like denials are are increasing. And then that second or, excuse me, that third line down, looking at reimbursement times and adjudication timelines, they are increasing. Right?

It's taking longer to get paid, and we felt through, you know, the early parts of twenty twenty four, right, many of you probably felt the payment delinquency from UHC, from AARP, from WellMed, some of these Optum carriers, and we've seen just that. You know, it's not as bad as that where you're going months without getting adjudicated claims, we've seen it go from, hey. We're getting paid in two to three weeks to it's now taking three to four weeks. And then just that little creep up on clean claims can be really impactful in a negative way into your revenue cycle and being able to properly report on, here are claims that we've billed out but have not gotten paid yet.

Those, we call those either delinquent claims or nonresponse claims, where eight thirty seven goes out and you, you know, you're generally waiting for two or three, maybe even four weeks for that claim to be paid. And in that two to four week time frame, you're not really concerned with the claim. You know it's gonna get paid or something's gonna come back. But once it hits maybe thirty days, all of sudden, that becomes really concerning that it's now delinquent. It's now been sitting with a payer for over a month, and it's now sitting in your open accounts receivable or your AR report, and it's moved in from that zero to thirty bucket to now thirty one to sixty, and you haven't gotten a remit yet. Those nonresponse claims can I can help identify issues that maybe the payer dropped the claim or is rejected, and there's no claim on file, or the payer just generally doesn't have it in their system? So that those all can look at as a driving factor into increasing reimbursement times.

So with probably just a couple more here in the industry overview, One thing I wanna talk about is isolating specific transaction sets and looking at if you're only reporting off of the remittance itself, so these eight thirty fives, the EOBs, you're missing a few key components to that puzzle. Right? I talked about the downcoding and recoding where you're comparing well, what was the build code versus what the adjudicated code, so the eight thirty seven versus the eight thirty five.

But a a ton of supplemental data that is on the eight thirty seven claim isn't on the eight thirty five. So certain things like diagnosis code and diagnosis code pointers, most of the patient demographics, Those aren't on remittances. Remittances are best for here is the allowable, Here is what the payer says they're going to pay, and here's what they're gonna push to patient responsibility. And if they're not paying it, here are the the denial reasons associated with that. Right? You've got your claim adjustment reason codes, your CARCs, your remittance advice remark codes, aka your RARCs.

Those things are gonna be pulled from the remit, but it's only once that transaction happens. And being able to pair those with an eight thirty seven claim and say, this is the first half of the equation, and I wanna be able to run analysis off of what our team is reporting as build claims. We'll talk about funnel mechanics here in a few minutes. Our team is reporting as build claims. Here's the second half of that remits, and it's tying those two together.

And the last little thing I wanna talk about from a reporting and really the impact of beginning to build your ROI business case around why is reimbursement is at every in q four, we're prepping up for CMS is gonna roll over in twenty twenty six. They're the twenty twenty six version of the physician RBRVS will get a new lab schedule, a new quarterly ASP drug schedule. Generally, about forty or fifty percent of payer reimbursement contracts use CMS guidelines as a base point or a basis for their reimbursement. So being able to report off of reimbursement history.

And some of the trends recently, I if anyone knows how the CMS physician fee schedule actually gets created, there's a handful of components that go go into it. First, you have all your productivity components. You have work RVU. You have a practice expense.

You have the malpractice or MP component, and that creates all of your total RVUs. But then each one of those also has a a geographic practice cost index or GPCI adjusts it based on your cost of living. Right? We talked about we've got all the way up in Alaska and all the way down in Florida.

I saw California, and I saw Massachusetts on on the early chats.

Like, all the different states and localities throughout the US, they have different geographic practice cost indexes.

And then all those total geographically adjusted RVUs get multiplied by something called the conversion factor. And you can see this trend in conversion factor from her total RVU in twenty twenty two is thirty four dollars and sixty cents. Now in twenty twenty five, it's currently at thirty two thirty five. And you kinda see how it stepped down over the last few years, which is decreasing reimbursement in many cases.

Right? There's a lot of provider specialties that their specialist codes are getting hit worse. Right? E and M codes have actually come up a little bit on the total RVUs, but a lot of imaging services, some surgical services, PT visits, a lot of those are going down in total RVUs.

And so between RVUs decreasing and the conversion factor decreasing, is sort of this double hit or double whammy on being able to report on your reimbursement. So I think of the decreasing reimbursement, a, driving more urgency around I need to collect as many dollars as possible to improve my net collection rate and using some historical reimbursement trends from both a budget standpoint as well as, hey. Potentially, If my if certain reimbursement baselines are going down, I need to counteract that by maybe contract negotiations or finding a way to increase my total revenue opportunity. So

that's some, you know, some things that we've just seen through benefits of being able to report on claims, some of the the driving urgency around increasing denials, decreasing reimbursement, and then we'll we'll just call, like, recodes and downcodes beginning to pick up momentum at least across four major insurance carriers that we've seen the documentation for. We're all like, Health care is kind of a unique special beast in reporting on claims.

How can we enable providers to see the most amount of data in a in a way that doesn't require a ton of, hey. You need to be a a PhD in data analysis to be able to get this yourself in a readily available, easy, as well as driving, here is the impact and here's the benefit right up front as opposed to digging for hours and go down a bunch of rabbit holes. Like, oh, we have a problem here. I wanna get in touch with that.

So moving to the second section, quantifying strategic impact. So thinking through what is what is our opportunity? Why do we invest? Why are claims so valuable for us in building or from building a business case.

So the first thing we look at is this, Has anyone ever heard of thing called a Sankey diagram? And it's a it's a really neat diagram that basically lays out where is everything going. Right? Where does all your money ultimately end up?

And so this is a an analytic that we put together that shows, right, you're gonna be capturing.

Here's everything that starts in a build status. Right? That's that first column right over here is everything that you send out the door could potentially be revenue. But then what happens with that revenue?

Well, you're gonna have, hopefully, the majority of your claims that are just gonna come back paid, and you don't have any issues with it. You're happy. You were paid cleanly. Then you have these buckets of potential issues.

One could be and we talked about denials.

Generally, generally, whether that's five percent, seven percent, I spoke with a group the other day that had a seventeen percent denial rate. There gonna be a big bucket of your your billed claims that end up in denials.

What about other issues that are maybe sneakier? Things like underpayments and overpayments. Right? If you negotiate contracts and you drive a multiyear escalator to your contracts or fee schedules or you recredential providers or add providers or maybe you have you joined a new plan or network, It's possible that you might not actually be paid according to your fee schedules.

And maybe you negotiated a five percent increase to one of your fee schedules, But that the effective date came and went, and you never got that five percent lift that you were expecting.

Those are underpayments, and those can be a a hidden source of revenue leakage. So this takes the build and says, happens once we get that first remit? Are we paid? Are we denied?

Is there a payment variance associated with it? And that takes all of those and says, well, what happens after that? Maybe some of my denials, I wanna bring them all the way up here to paid. For my underpayments, I can come through and Recover those underpayments so they're paid correctly.

So you just follow the life cycle of each one of these. And some of the benefits of this is I think, that second bullet is probably the biggest one, you can identify the initial revenue leakage points. Do you have a problem with payment accuracy and auditing correct payments? Do you have a denials issue?

From a denial standpoint, how good are you at recovering once denied claims? How efficient is your your AR follow-up team at resolution? And you can begin these getting these KPIs. And then from a secondary issue is, you know, what a lot of times will come through and find, hey.

This was denied, and that's the primary issue. You get things appealed or you send a corrected claim, and then they adjudicate and pay it.

But when they pay it, they don't pay it correctly. Those are underpayments. So working through these Sanki diagrams allows you to see, well, it was denied. And then from the denial, it came back and it was underpaid, or maybe it was redenied.

Why why are these secondary issues populating. And to put this into perspective of a real business case, Here's what we so funnel metrics, just about everything that you do, ultimately, is a funnel. Right? And it is it starts with how much revenue you're capturing. And there is some leakage oftentimes with groups. There's some front end leakage on everything that the providers are serving and creating from a potential revenue opportunity, like, generally, you try and capture a hundred percent of that. Let's say that you're leaving just one percent of that on the table.

And then from that revenue capture, the next process or next milestone in the process is we 've coded it, and we've generated a claim, and a claim is now going out the door to the payer. Right. So imagine that you're just losing one percent of that through, we call it, discharge not final build. So that's a, you know, second leakage point.

And then you look at, hey. What's a pretty average denial rate? And let's say about seven percent is a pretty average denial rate in the industry. So you take that revenue, and you're looking at almost three hundred thousand dollars of initial denials.

And then what's your opportunity from a resubmission?

So about, again, on average, about sixty percent of denials are successfully appealed, rebuild. So what's happening to the other forty percent? Are they just being written off. We call those failure to resubmit. So in each kind of section of this funnel, you see where the revenue opportunity is and where the those leakage points begin for each one of those. So it's helpful to maybe put some real dollars behind a a practice that, you know, that does ten million dollars of revenue a year type of thing.

So getting into a little bit of categorization, Right? We'll kinda go this front, middle, back, and You know, starting on the front end side of things because so many denials actually stem from front end process breakdowns. Right? So we we go through and in our time, we've we've accessed about half a billion claims and remits through the Rivet platform and begin to take the claim adjustment reason codes, and we assign them back to functional area. Right? So things like registration and eligibility.

Maybe that could be bad demographic collection. It could be the patient isn't actually eligible, and maybe for some reason, they slipped through our automated eligibility process to verify coverage. Right? That is the number one denial across everything, but especially on the front end, is registration and eligibility related styles.

And then you come through what else is tied to that. Coordination of benefit denials, that's, you know, that's a big one. There's no very easy right answer oftentimes on coordination of benefits other than verifying with the payer that they are listed as primary and this is listed as secondary.

So that's, you know, that's another big one. And then one of the hard ones that has a high fatality rate in terms of write offs, you see this section right here for authorizations. So about fourteen percent of all denials are related to prior auth, referrals, precertifications, and those ones are are typically yeah. There's a difference between fatal and nonfatal denials. Nonfatal, typically, you have a successful appeal process, and you can get those paid. Fatal denials end up in write offs where you get zero revenue and the balance is adjusted off.

It looks like Fire off. Okay. Andrew, I was just about to say, looked like you froze there for a bit, but you're back.

Oh, alright. I I don't know how much. Well, thanks, Lee, for for stepping in and but the the authorizations can be very detrimental because they have a high fatality rate on the from a recovery standpoint. It's very difficult, especially as outpatient providers, to get retro authorizations. So think through that's some of the pain points that we see derived from claims is the front end process breakdowns that that drive denials coming through. And about forty percent of all denials stem from front end.

Then we come to your middle. I we've HIM coding, few different names for the same types of teams is there's three leakage points that we look at in HIM.

There's charge lag, charge capture, and revenue integrity.

So the charge lag is how long it takes from an encounter to be finalized to when you're getting charges ready to go out the door. Right? And if you have a an unnecessarily high charge lag, generally, providers should be submitting their charges, and then they should be get coded up, you know, within one to two business days. But if you have a higher charge lag, that's going to lead to slower claim submission and ultimately higher accounts receivable.

From From a charge capture standpoint, it's, are you getting all the revenue, and are you billing for all the services that were actually rendered? Now this tends to be oftentimes a little more painful on the facility side of things because there's so much that goes into an inpatient claim, for example. And you may think, oh, it's just a claim level DRG payment, but you have all these ancillary services like drugs and biologics that may get carved out. But on the prophy side of things, are you billing for the correct units?

Are you billing for everything that was actually rendered? Are you billing for the correct level of e and m care?

What we consider kinda charge capture and revenue integrity is, are you capturing everything that ultimately should be done?

And From there, you You look at let's say that you've gone through, you've done any everything clinically necessary for a level four code, A level four e and m code. Here's a little callback to that one again as just, like, why this is so helpful is if your team is your provider, get all the requirements for a certain level of e and m, Your coding team agreed with that, entered the charges, submitted it, and then ultimately got paid thirty percent less.

These can be really painful stemming from not really an HIM issue unless you truly are overcoding things that are not medically necessary.

But most of the times, especially when we've seen appeals go through these and they have a really high successful success rate of appealing those down codes, is just unnecessary work.

And claims are ultimately being impacted by by that. Right? So we we flag one of these things called the recode, and you can see how big of a deal those are, you know, just some volume, is if you look at a recode and where you're paid less than what you should be getting paid, most systems are just passing these through as a zero balance clean clay pay clean paid claim because there's a contractual adjustment on the claim or on the remit and say, cool. We have a charge.

We have an adjustment. We have an allowable. We have a payer paid. It's just thirty percent less.

So if if you're curious about these, you wanna know what they actually look like.

This is how they would show if you have sort of a ledger. I know a lot of EMR and PM systems have, effectively, like, a ledger that shows, oh, here's the charges that were submitted on the claim. Here's the the line items from the remittance. And this is just our, you know, our visualization of it. You can see that this service has a nine nine two one four with a twenty five modifier on the claim. Right? See, this is our outbound eight thirty seven.

The remit comes back and say, well, instead of a nine nine two one four, the payer just autonomously changed that to a nine nine two one three, and they never really told you. Right? Sometimes there's a claim adjustment reason code. Sometimes there's a remittance advice from mark code that'll tell you they're downcoding.

Oh, I think that is I think it's c o one eighty six is the adjustment code that you could look for or an n twenty two remark code. Both of those tend to indicate that the payer changed the the code they adjudicated. And then recently, though, payers aren't even putting those on the remits. They're just putting a c o forty five contractual adjustment saying, nope.

This is what we're paying. So easy way to slip through revenue.

And the last two, I wanna talk about one, timing of accounts receivable, and two, we call it the double whammy.

So the the timing is we look at this from we we segment out claims that are paid clean on the first pass. So few different ways that you can you've probably heard this named.

One could be your, your first pass rate, meaning that you submit a claim, it gets paid, it goes to zero balance, and you're happy.

The other segment is claims that have issues with them. And so dating back and looking through the entire claim library from our twenty twenty four database, right, so January through December of twenty twenty four, On average, claims that were cleaned that were paid correctly the first time, that took about sixteen days to get adjudicated.

However, claims that required follow-up or supplemental action, that jumped from sixteen days to ninety two days. So, So, like, well, where is our accounts receivable coming from?

Well, it's gonna be hard to get it less than sixteen days on your AR because that's just a standard adjudication cycle.

But if you're steering down thirty five, forty, forty five days in accounts receivable, it's this bucket right here, which is how quickly is your team able to once something bad happens, how quickly is the team able to remedy that to identify that there's an outstanding issue? And how efficient are they in recovering that revenue. And it's gonna take right? If you can get to if this is sixteen days, if this could be two to three x, this average time line, That's like, you're doing really well on that.

But you can see that would be what's that about? Forty eight days. Anywhere from thirty two to forty eight days would be a a good range for follow-up claims. Ninety two days is really long.

That's meaning that not only are you going through the average adjudication cycle of sixteen days, but then it's wait sitting on your accounts receivable to be prioritized. Maybe it sits on your AR for a week or two until your team's able to get to that specific claim and resolve that. And then payer readjudication is always gonna take longer. Right?

If they take on average sixteen days to pay on the first pass, you can generally it's going to be probably close to double that on appeal reconsiderations and corrected claims.

So that's what's driving that that ninety two days.

And then, ultimately, on the very back end of things, We call this a double whammy and it's write offs.

Write offs are so painful, especially if if you're involved in any of your provider compensation reporting.

There's while you may have if you're a small group, it's likely you probably have some cash based payments your providers just paid a percentage of what was collected.

If you're bigger organization and you recruit providers, there's probably some component of it where they're on a salary, a draw, or a productivity based compensation. Right? We've seen a lot more groups are evolving into this hybrid of both productivity and quality or productivity and cash.

But all the productivity components, there's a write off, it just you go through that funnel and say, Well, we spent our our, physician time or provider time rendering services. There's a cost to that. There's a facility cost. Right? They're going to a clinic room or an operating room or an imaging device.

That is overhead cost. And there's an opportunity cost to by not getting paid for those services, where you could have different patient, their service may have been covered. So there's hard costs. There's provider costs. There's nonclinical FTE costs to try and resolve these denied claims.

And in some cases, there's certain like, physical costs or maybe it's drugs, implants, devices.

I was wondering with the group just recently that every time they do a specific surgical case, the device cost is seven seventy five hundred dollars. They pay that the seventy five hundred dollars to a DME vendor. And then if they don't get anything on that service, not only are they out the opportunity cost of the revenue, but they're out seven thousand five hundred dollars for the device they used in that surgical service. So there's a ton of costs that are born, and then, ultimately, you're getting zero dollars for revenue.

Like, that is that is really painful to offload all those costs or pay all those costs and ultimately get no revenue on the back end. So that's a double whammy of write offs. So that's why I think if you can if, a, there's anything you can do from a prevention standpoint and then successfully appealing things and determining using claims analysis why what's our recovery rate? Where are we not recovering things? If we're not successfully recovering certain types of denials, what can we do to layer in prevention stop those in the on the front end side of things? We'll talk about more of that in a moment.

That is gonna pay off dividends on reducing write offs, increasing your net revenue collection rate or net collection rate, and ultimately decreasing accounts receivable and increasing cash.

So that's it for kind of our second section.

I then wanna walk through a few of the the key analysis components that we look at, starting on the revenue generation side of things. So how to analyze and initiate some of your changes. So if you're familiar with the relationship between gross and net revenue, gross revenue is it's just your build charges times the the quantity or the units you're submitting.

It's for all intents and purposes, especially given that most of you are on the prophy and ambulatory side, it's funny money in the sense that if your contract rate is a hundred dollars, it doesn't matter if you bill a hundred and one dollars or five hundred dollars. You're still getting paid a hundred bucks. With the one little asterisk that if your contractor rate is a hundred and you bill ninety dollars, because most of your contracts have lesser of clause, you're gonna be leaving ten dollars on the table every single time.

But it's helpful from a a forecasting standpoint to at least know kinda where are we trending, what's our productivity. So from a revenue generation side standpoint, we look at that top, here's your gross revenue. And then if you can forecast out your net revenue or allowable is going to be, that can be really beneficial to begin to measure against that net revenue forecast, the cash that you're collecting. Alright.

So pretty deep into the year, you can see our January twenty twenty five cohort of services, that's largely been adjudicated and paid at that point. I bet you're you've already received, at least from a payer revenue standpoint and a a net revenue or allowable, probably ninety eight to ninety nine percent of the stuff you're going to receive has already been adjudicated from January twenty twenty five. Same thing with February, March. As you move into more, excuse me, more recent months, you You'll see that the gap between this little flashed purple bar and the purple and orange bars begins to increase.

And that's because that is the delta between what your system should forecast as your net revenue, aka, like, is everything that you could collect, and what's actually been remitted and received so far. So the reason that is so helpful is without a forecasted net revenue number, you never really know when you've collected everything that you should collect. So the KPI that's shared is you should be driving towards a ninety five percent plus net collection rate, meaning of all the revenue that you possibly could receive. Imagine no denials, no write offs, Every patient paid all their dollars every single time.

Hey. I don't think Lee and I would be on this webinar because you guys wouldn't have any problems. But b, Like, you would have a perfect net collection rate, And that is the most contractual like, based off of your contracts, the most revenue you could ever receive.

A step down from that, if you can lose less than five percent between bad debt write offs with patients, insurance write offs for denied claims, your, you know, your net collection rate is fairly strong. But you need to know what is your actual net revenue in the first place. That way, you can measure against that and layer in those collections.

From a trending looking coming back to this denials, we like to look at, not only just that pie chart of where denials are coming from, but also trending it over time. Because there's a lot of times, you'll see that you're going along with sort of flattened denial trend in history, and then you'll see a big spike. Like, woah. What happened in that time period?

We need to assess, like, why did providers start getting these denials? All of sudden, we're getting these weird taxonomy denials. Or on October first, this payer started sending a bunch of c o sixteen denials requesting supplemental information. Right?

It's helpful to have that trend over time so you can see if there's a a peak somewhere on a specific root cause of denial and say, well, is it missing or invalid data? Is it coordination of benefits? Help me understand where our trends and spikes are coming from.

And then you can also drill into provider level details.

Alright? So we take the you can look at revenue. You can look at productivity. You can look at denials.

When you begin to segment and break that down by Excuse me. By provider, you can see where are issues stemming from. Do we have if you're a multispecialty group, you may have your primary care and your pediatricians, and they have maybe a three percent denial rate. But then you've got some of your specialists, and they're at a seven percent or eight percent denial rate. And you can look at the residual cost to collect on a specialist versus the primary care provider that typically has less revenue per provider, less denials per provider, and compare that to some of your specialists, woah. We have a real issue with our orthopedic providers or our ophthalmology providers.

Or if you're a single specialty, maybe you're just ophthalmology, and you wanna compare similar providers across each other. So you've got a cataract group, and you have ten cataract surgeons. He's like, well, most of our you know, nine of our providers seem to be in the same bandwidth of productivity, of RVUs. They're producing the same amount of revenue. They have consistent denial rates.

But then we have this one provider, and that one provider is the outlier in terms of revenue production or denials or resolution or net revenue.

You can look and see and be able to compare and contrast provider analytics relative to each other.

So it's helpful to see providers. And then we talked about this a little bit already, but it's your maturation of revenue curve. So So if as soon as you generate claims, you now have actual revenue produced with those. But from a budget standpoint, how can you look into the future?

And one of the best ways to do this is accounting for seasonality, but beginning to look at, well, what is your average daily revenue or ADR? This is the if you're familiar with the AR days calculation, right, it's like, oh, we're at forty days in AR. The average daily revenue is the denominator associated with that. So you take your total accounts receivable divided by your average daily revenue, and that gets you your AR days.

You can also use that average daily revenue to project into the future What should you know, you can look here in the future, be like, woah. In January and February of twenty twenty six, clearly, we don't have any claims yet, but I wanna understand what what do we believe from a budgetary standpoint we should be producing so long as we're keeping up with normal productivity.

And, you know, for what it's worth, January is a very long business month usually, followed by a very weird short month. It's like the the juxtaposition of January and March with February right in between is very weird from a cash and a cash posting standpoint, but you can project out what your revenue should be and how you're collecting against that revenue and how quickly. Right? So if this cash curve gets steeper, it means you're just collecting more revenue faster.

And we talked about that cash to net. Ideally, you should at least be at ninety five percent. Seen some some industry leaders being over ninety seven percent, and they're just really efficient. They're They're implementing front end processes for patient balances, like patient cost estimates to collect on high dollar deductibles and out of pockets.

But if you can get ninety five percent is, like, that's what you should be shooting for, but we have seen ninety seven percent plus from a net collection rate.

So coming back to the analytics component a little bit.

Zooming out from the the weeds of some of the problems you've been talking about, I wanna talk about just the premise and and good data.

We think of the the four a's of analytics. And they're accurate. They're automated, available, and actionable. So accurate, obviously, if you're producing bad data in, bad data out, bad data out means bad, you know, bad initiatives, bad planning, bad strategic planning. So if you if you can't trust the data, you can't act on it, or if you do trust it and you act on it, you might be going down the wrong path.

So accuracy is very paramount, and that's as best as you can pull from all of your systems, your claims data, your remit data, your charge data, And then the automation side of things.

This question to ask yourself is, every time you want a report or you wanna see certain data, you have to go through, like, a SQL analyst or the IT team, or you have to submit a new report request to your local health system that you're on their system or borrowing their EMR or PM system. There's no way automated. And your ability to iterate through analytics is definitely hindered. Availability is and we we think of anyone who could potentially be a key stakeholder in the process, the outcome, the revenue, they should have access to these metrics. And they should be able to, from an action actionability standpoint, guide you on a path resolution. Right? Data is really easy to produce.

Good discovery and good storytelling with data is a lot is, you know, night and day different. So data just is going to lead you down a lot of rabbit holes of, like, well, I'm gonna try and discover something. But being able to articulate and have your data tell the story around where is your biggest opportunity, that's kinda next level. And that's where we're seeing some of the industry trends with AI being able to auto generate anomalies from your data and pull that in and say, this is a trend that previously was not a problem.

We're seeing that becoming more and more prevalent. It's pretty neat stuff to be able to effectively have an AI analyst that you can ask it plain text questions and have a conversation, and they query your data and pull back results. So the the automation, the availability, and the actionability of your analytics gets is is pretty neat.

So where we go from beginning to prepare your data, This might get a little into the technical side of things, but looking at cleaning and transforming your data. We talked about how health care is so weird and complex. There's the EMR PM dataset. There's the claim eight thirty seven dataset.

There's the remit eight thirty five dataset. Those are three components that, excuse me, that all are in factor or get factored into a good revenue cycle analytics platform. So I think about what data do you wanna report on? Where does that data come from?

And what data element produces that? Is it coming from claims, and you wanna be able to track This value based care metric or this quality incentive or Yeah. These specific measures. Right? You have to start with good, clean data, and you have to define what you want in scope for your report.

And then from there, you begin to identify patterns and trends. Right? So while I am I am no sewer, those look like some very exciting clothing patterns.

We talked about some of the AI components that will whether it's LLM or large language models or some of the generative AI around claim discovery or just statistical techniques that have been around for a long time, like regression analysis.

Those will help you synthesize some of the data in a in a pretty straightforward way and help you save a ton of time in pulling out trends or takeaways from some of your data. So there's a lot of leverage that you can pull from modern AI, whether that is the generative AI or agentic AI that is going to autonomously query your mind through a ton of your claims, or you wanna interact with it in a way like, if anyone's used ChatGPT and you'd say, hey. I want you to create this. Like, just recently used to create my eight year old daughter's birthday invites.

We found a template. We upload it and say, can you customize this? And it just generates it. Right?

That sort of interaction is really neat when you see that in the context of data.

And last but not least, visualizing your findings.

Data is often really overwhelming, especially at the scale. Right? If you're talking about an average provider is maybe averaging about three thousand encounters a year for not even that productive of provider, then let's amplify that into how many codes are they billing. On average, they're probably seeing two and a half to three codes per claim.

And so you you begin to build this data array. Like, well, how could you visualize a hundred thousand data points? It's not helpful. So transforming that data from raw data points into a trend, into a distribution, into an area graph.

It can be really helpful to be able to see the the overview aggregate, but then being able to drill down.

This is where going down that rabbit hole can actually be really helpful and valuable is once you see an anomaly or a trend or an outlier, you kinda double click into that bullet point and say, what happened here? Oh, here's the claims that are driving that analysis.

Let's go and do a root cause assessment for that batch of claims to determine why are we getting to this point.

So visualization can be really helpful, especially visualizing it the right way. Right? Time studies, yeah, those are best demonstrated in line graphs. Area charts or pies are good for distribution graphs. If you have multidimensional, you may be looking at something like a scatter plot. And really good analytics and BI platforms should be able to push put this data in a dimension that allows you to quickly see the results that you're looking for.

So that is some of the the components of analytics that we think are helpful just to keep in the back of your mind as you're as you may be reevaluating, what is your your rev cycle? How are you producing them, where is the data coming from, what systems are you using it, who is ultimately responsible for this, all that could be really helpful. So wrapping up with a couple minutes left here, shifting into some of the actionable takeaways.

I'll do a a quick q and a check here.

The The applying your analysis is, ultimately, what are your goals? And I I bubble down to analytics can benefit, you know, many, many ways. But one, you can improve Excuse me. Your operational efficiency.

And that that can mean a lot of different things. Right? Are you looking at front end throughput? Are you looking at claim submission process?

Are you looking at how efficient your your cash posting or your denial resolution and appeals process is?

Terminating what you right. Starting with the where is our problem stemming from? Where is our leakage point? Then using the data to drive operational efficiency throughout your revenue cycle. They can also help you manage some of your costs. Right? If you have a a high cost to collect or if you have certain fixed or hard costs is is you can quickly see into this some of that from a a cost standpoint.

And then looking at forecasting or strategizing is where do you want to go as a business? And historical claims will say where you've been, and they'll help you predict if nothing changes, here's what you're gonna look like in the future. I health care kinda demands this ability to sort of see into the future a little bit. A, it's not an industry that you're gonna get paid overnight on, so you have to be able to see weeks and months out in the future because of the claim process.

And then from a a prioritization standpoint is where are you gonna be putting your resources? What are you gonna do with resource allocation or strategic planning? What what should be happening six months from now, twelve months from now? And that's gonna drive where you're deploying resources and assets from a, you know, an an OpEx and a capital expenditure standpoint.

Looking at the future projections of your revenue, So there's a little recap. That's it. Eight. It's good to know what you should be planning. Because if you see certain claims that are or a low month of claim volume or maybe you're going through a system shift where you because you went on a new clearinghouse in a PM system, you held claims for three or four weeks.

Be able to articulate out what that cat residual cash impact is gonna be three to four weeks later based on normal adjudication cycles. You know, it's just helpful to see that and helpful to know. Some Some of the other data that I think is helpful is external benchmarking.

Yeah. Know, we we talked about net collection rate and the initial denial rate, but also some of the larger data models that are out there. Know, if you are familiar with the transparency and coverage data files. And you wanna know how do my rates for my commercially contracted payers, how do they compare to rates of other providers in my network in my market?

Right? Transparency and coverage data, it's huge. Right? You it's not something you just like, oh, I'm just gonna drop this into Excel.

It's terabytes and terabytes and terabytes of data. I think the last time I heard this metric was riveted indexed eighty two trillion rates through the transparency and coverage data files. Massive amounts of data that that allow you to see how your rates compare to those in in your market based off of taxonomy, specialty, key provider comparisons. So those are some other external benchmarks that you could use from a data standpoint to say, how is my reimbursement?

And then last couple here, I'm asking of the the KPIs that are most impacted by claims analysis. So the claim denial rate and clean claim rate are both Directly derived from what is your claim throughput process. Right? If you are getting claims out the door as quickly as possible without billing edits, without scrubbing them, without stopping them and looking for potential rejections or denials, you're gonna impact your denial rate and your clean claim rate, which in turn is going to impact your days in accounts receivable.

You So just Sarah's question on the the days in AR equation.

Right, that is the total accounts receivable balance. So say you have ten million dollars on your accounts receivable, whether it's an open AR report or some groups call it an aged trial balance or an ATB.

The taking that as the numerator. So let's say it's ten million dollars or a million dollars divided by what your average daily revenue is produced at. And your average daily revenue, depending on You have gross AR and net AR?

I didn't know the most organizations leave their accounts receivable on a gross basis, Meaning that when you bill out a hundred dollar claim, it sits at a hundred dollars until adjudication happens versus there are some groups that have net down or proration at the time of billing. So that hundred dollars, when you build a claim, the system says, oh, you're gonna get sixty dollars for this. Your AR is now sixty dollars. It's just it's very complex.

Most groups don't have that sort of methodology built. So your gross AR is sits at a hundred dollars until adjudication comes back. So your average daily revenue is looking at your total AR divided by your average charges produced in any given day. That's your accounts receivable collect formula for AR days.

And then if your AR days are really high, you've got high balances, your team's gotta go through everything, That's gonna impact your net collection rate because, ultimately, you're never gonna be able to get through your entire accounts receivable list. It's just too much. There's too much volume that balloons out, so you can't get to it. So you end up writing things off.

You lose net collection or you impact net collections. And as you're trying to work through this entire list of of open claims, Maybe you're touching two things two, three, four, five times to get paid. That's a really high cost to collect. Every touch that you have on the between claims has a as a hard cost to it.

So if you're touching it and you have a low efficiency, three, four, five x to resolve an open denial or an open claim, those are all impacting your cost collect, and then, you know, just trickles down through the rest of your funnel of your KPI metrics.

The last but not least is identifying opportunities to increase your revenue. There's really only two ways that you can increase your revenue.

And, you know, we're we're getting we're getting close on Thanksgiving time. So imagine this is a nice revenue pumpkin pie. Is you can either increase your total collectible revenue, and that's maybe you're adding providers. You're adding services. You're changing your procedure mix to drive higher dollar procedures.

Or Excuse me. Or maybe you're making every every procedure a little more you negotiate the reimbursement, so it's a little more cash per encounter through negotiating insurance contracts. Those are the ways that you would increase the size of the pie. You're not actually getting more efficient or better in any of your operations. You just now have a bigger pie to collect ninety five, ninety seven percent of, Or you can increase your slice of the pie that you're collecting through looking at Preventing denials by comparing, hey. What are we what are we fixing on the back end? How can we resolve these?

Prevent them from getting denied in the first place. How can we make sure that we have really high efficiency, and we're reducing the amount of touches to collect on open denials and ultimately writing off less revenue and driving from maybe you're at ninety to ninety three percent net collection rate right now, and you wanna drive towards a ninety five, ninety seven plus percent, that's gonna increase your slice of the pie.

So with that, I just got a monologue for the last fifty five minutes or so.

We've got a few questions already dialed or already sent here. I answered a couple questions.

I I guess Yeah. One actually on AI, and I thought that was a good one. I think someone was asking about AI and the impact of AI on data analytics or if we thought that AI would take over analytics. And I actually think that is a great area to leverage AI.

Just the sheer volume of data, as you indicated, and having to sift through all that data, a lot of challenges that organizations face is resource. And the reason why they can't get a good handle on the revenue cycle is just lack of resources across many areas.

So why not use AI to go through some of that analytics and save those resources to actually act upon that data output that the AI is able to provide.

So, I mean, I personally think AI is a great area to incorporate when it comes to data analytics and flushing out some of the data so that you can act on it.

Yeah. Yeah. Definitely, Lee. And and piggyback on to we had a question from or just a comment from Sherry around, hey.

Watch what AI is giving you. When you train a new a AI model, it it is going to learn from a set of data. And now one of the ways that Yep. Yeah.

I'll just speak to kind of Ribbit in this sense because all of our AI is is proprietary trained to a customer's account is it's loading in your claims, your remits, your, you know, data that gets presented by your specific entity in practice as opposed to some generic claims database. Right? That is the generic claims database. It could be helpful if you're talking about industry high level trends or KPIs and you just wanna know general results.

But your group is going to have your specific claims, your CPT codes, your modifiers, your providers, your denials that are specific to your organization. So a strong AI model should be proprietarily trained to your specific claims data so that it can provide results that are highly contextualized to your organization. Right? If you're an orthopedic group and you're using an AI model that's trained on a total claims database and they're saying, hey.

Watch out for ophthalmology denials. Like, that's not helpful. But if it's learning from, you know, a year plus worth of your claims data, so they have cyclicality, they see denial trends, they have a sufficient payer mix and procedure mix, that can all be really helpful to drive accuracy in the AI. And then I I think as you're getting started, the trust but verify is certainly true, as especially as you're getting ramped up and you'd be like, alright.

We interact with this, and they presented this as some of our findings.

The first iterations through, you might want to feel comfortable running some parallel analysis through that so that you can kinda yeah. We're we're coming up with the same results, the same definition. And then what's really neat about some of the like, borderlines on some of the generative AI is the ability to dynamically tweak their definitions. So things like I'll I'll share an example of one thing that was getting classified as a bunch of denials.

Right? If if you're all familiar with secondary payers, they put a lot of o a twenty three adjustment codes on secondary remittances. Basically saying, oh, payer paid this. We're not gonna pay anything else.

In many cases, there is no problem with that. But this is you know, an an untrained AI model might come through and say, woah. You have a ton of secondary denials. We're not paying anything.

Reality, there's no revenue that you're gonna capture there. So you can, like, literally through text train that model to to dynamically update its definition for things like secondary claims. So it's really neat how you can evolve that.

Yeah. And and last little couple here.

I know we're we're short on time that Is there any help for a professional that has started career in medical billing and coding?

Yeah. I mean, AAPC, they do tons of content, both CEU and and non CEU.

Yep. We overseas tons of educational components here with AAPC, like, tons of resources, tons of formal certifications. If you're on the the coding side of things, like, official coding certifications, you're on the billing revenue cycle, tons of content provided through AAPC. And, you know, if if you're as part of a specialist, there's a lot of forums that also can provide some good opportunities that are are specialized to your your specialty. So add that as tons of industry resources. AAPC, one of the kind of industry leaders in in content knowledge certifications, proprietary data models, things like that.

What was the average target percent for bad debt?

Yeah. I think There's I would add in bad debt and insurance write offs together and say those are if you can keep them less than five percent, you're doing okay. And whether it's three and two, two and three, four and one, just trying to get all of your write offs ultimately lost revenue as one pot and try and keep all of those sub five percent. That's that's how I think about the bad debt percentage.

So I know there are maybe a couple other questions you didn't get a chance to to get to, but wanna be respectful of of all of your time, and and thank you so much for hopping on this AAPC webinar and hanging out with us and with me and Lee for the last hour or so. Really appreciate it.

Thank you both Andrew and Lee, and thank you to everyone in attendance for that lively chat and very supportive of the community.

Reminder, we're gonna because I know there are a few questions about that we wanted to have answered about previous slides. So that's my cue to remind everyone that you'll receive a copy of the slides and recording of this presentation. I think that should be in your inboxes on Monday.

And if you are interested in learning a little bit more about Andrew and Rivet, there's gonna be a survey that pops up when we end this webinar that you can indicate so. Thanks again, everyone, and we hope to see you next time.

Thanks.