This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong

Keynote: Exploring How Emory Cut Denials in Half with Carla Haack, MD

Carla Haack: Everybody wants to do the right thing and everybody wants to do a good job. But despite the fact that we work within the same system, we live in such different worlds that it's really difficult for people to understand each other.

Bill Russell: My name is Bill Russell. I'm a former health system, CIO, and creator of this Week Health, where our mission is to transform healthcare one connection at a time. Welcome to the 2 29 Podcast where we continue the conversations happening at our events with the leaders who are shaping healthcare.

Let's jump into today's conversation.

Alright. Welcome to the 2 29 podcast. Today I am joined by Carla Hack, Dr. Carla Hack with Emory Healthcare out of Atlanta. And uh, she is the Chief Financial Informatics Officer. And we're gonna talk about that because it's very distinct. Carla, welcome to the show.

Carla Haack: Thank you so much for having me. It's an honor.

Carla Haack: Fair enough. So the role is, is very exciting and very fun for me, and it mainly consists of being a facilitator of multidisciplinary collaboration in the pursuit of helping everybody in the organization have a better day and allowing us to fulfill our mission of serving the community, improving lives, and providing hope more effectively.

Bill Russell: You know, so you multidisciplinary, you're sort of a somewhat of a liaison between finance, IT, and clinical

Carla Haack: .Perfectly described, and,

Bill Russell: and you identify those areas where there are.

Inefficiencies, and it could be in a lot of different ways, but a lot of them happen to be directly tied to reimbursement and physicians getting paid and the system getting paid. And they're not small numbers.

Leakage in your revenue cycle. The way to make that up is to be more productive on the end of the revenue generating team members. And when you're at the end of the pandemic, I had done emergency general surgery throughout the pandemic and we had stayed busy, but all healthcare systems were struggling financially.

And there was a message sent to clinicians that we weren't productive enough and acknowledging that we had shut down elective surgeries.

Bill Russell: How, how, how is How was that received?

Carla Haack: It was a bitter pill to swallow. I will not lie. And I'm not, it is not a criticism of our leadership at that time, because if you don't know, you don't know.

Bill Russell: Right.

Be where I needed to be from a family perspective. That ultimately led me to having bandwidth during hours that would normally be clinical and operative prime time. That allowed me to show up to revenue cycle process improvement meetings that I would've never gone to if I'd been in the operating room or the clinic.

So I start showing up to these meetings and I start analyzing denials. I realized that people are assuming things about the mechanism of a denial because they cannot interpret the clinical documentation to figure out whether what we put on the claim is what we said we were gonna do when we saw the patient in the office, and whether that's what we actually did when we went to the operating room.

So I first really started sliding down this slope when I started becoming a translator for clinical documentation for people who were analyzing denials in the revenue cycle. And just realizing that by helping people work together and understand each other, we could really make a big difference. And the digital team is where the rubber meets the road.

Bill Russell: So you've given me a lot to go off of here. So the first thing to note is, uh, in a lot of these cases, the work was being done, the clinicians were doing the work, and still you had it goes through the process and the clinician ends up not getting credit for doing that work because it's not that something breaks in the process.

Carla Haack: If I could clarify one thing,

Bill Russell: please.

Carla Haack: At our shop, and I can't speak for other major academic centers, but at our shop, the clinician gets the RVs independently of whether or not the system gets paid.

Ah, and I have interesting feelings about that because on the one hand, the work has been done, the stress has been felt, the effort has been made. And yet, on the other hand, if the system doesn't get reimbursed for all of that work and that effort, then it does not, it's not sustainable. And we did historically.

Bill Russell: Interesting. I want to take you back and, and this is more of a, a case study. because I will tell you that when in our 2 29 meetings you have, you have, 15 CIOs sitting in that room. Alistair would stand up and he would tell some of the stories of the amount of revenue that was captured in the amount of, uh, you know, in the things.

And he would talk a lot about the fact that having someone focused on this, your role is very focused on this, this intersection that a lot of times it's just broken there. There's just, and I'm gonna get to that in a minute, but he would talk about this, and I'll tell you the 14 other CIOs would be sitting in the room taking those furiously going, I wonder what our percentage is.

Who's in those meetings? I'm curious. I'm back in the day. I'm, it's probably different today, but back in the day, I, because what I'm picturing is a silo. I'm picturing a bunch of people trying to figure something out and they're the right people aren't in there to help them figure it out.

Carla Haack: You are spot on.

So you admit a patient, you make them inpatient and the insurance company comes back and says, no, that's observation doesn't meet medical necessity criteria for inpatient. And I would do a lot of those peer-to-peers with the insurance company. That started sort of opening my eyes a little bit to how it was that the insurance companies worked, et cetera.

Right. So if you start at the end, you are not actually able to fix the root cause of the problem. So the denials value stream analysis sort of ended up with, okay, well we need a, we actually need a pre-cert value stream analysis because that's where this all starts, and that's where a lot of these denials starts.

So that was then the rep cycle process improvement work that I started getting involved with. After the pandemic, and I'll give you a really simple example of things that would happen. Sometimes you didn't have an off, and when you don't have an off, well that's easy. You don't have a leg to stand on you.

And there was little to no reconciliation happening in between, and they would assume people who were looking at those denials would assume that it's because the surgeon did something different in the operating room and didn't tell anybody.

When we started looking at those cases one by one, there is no glamorous way to do it.

The way that we, and the way we started doing it is that we had a bunch of rev cycle team members in the room, and we had a bunch of denials people in the room, and they would look at these denials and they were having these conversations. I was like, wait a minute wait. Excuse me. Sorry. Stupid question.

And there were a bunch of different reasons why the codes could be different. You could have posted it wrong, you could actually have your system built wrong. There were situations where we would. Order a study. Or a service and the CPT code would be attached to that order, and we kept getting denied. And when we looked at the CPT code, it was the wrong CPT code attached to the order.

Bill Russell: People forget, I mean, when you're doing those builds, in a lot of cases you're doing a lot of work at a, in a very compressed time, and some of that stuff just does, there's just errors that are made.

Carla Haack: It's. Easy to fat finger something. It's very easy to make a mistake

because the way that I think about this is how do I set our digital infrastructure up such that the work that happens during the course of clinical care translates efficiently and effectively downstream to the revenue cycle with. Out making our clinical teams do anything that they're not already supposed to be doing for quality and safety purposes.

So what we did, and this took a while to stand up, but what we've ultimately ended up doing, and I'm thrilled, and it's it went live on May 21st, so it's still relatively new and it's still my little baby, is that we've partnered with our quality team who was doing a reinvigoration of our universal protocol, which refers to the series of things that you do in the operating room for quality and safety purposes.

That is a magical moment to actually capture what you did and put it into the system in a way that allows us to notify the payer if what we did is different than what we had authorized. That's how we set our system up. We partnered with the quality team. We told everybody, Hey, you're debriefing. You're supposed to be debriefing.

Everybody needs to debrief. This is a quality and a safety thing, and we will not compromise on that. We're also going to supercharge your efforts to make it so that what you say at the time of debrief actually helps you get credit for the work that you have done. And that is, those are important words because it's really important to make sure that everybody knows that we're doing this, so that credit is given where it is due.

And we changed the field in the intraoperative record, which is the, the document that the circulating nurse documents on. And if there is a change in the codes in that procedure field, those codes auto-populate based on what was pre certed. And if we, if we put something different in that field, that account automatically routes to a work queue.

That's worked by a nurse that's, that works for our central precert department, and she submits those cases to the insurance company on a same day, next day basis. The insurance companies will usually give you about 24 hours to submit a post service request where you say, I did this operation. I precert this other thing, but this is what I actually did and this is what I need authorization for, and it took a million dollar bite out of that particular denial mechanism in the first 21 days of being live.

Bill Russell: there is no easy button for this. Is there? I I think people hear, oh, I heard a vendor who said, we're gonna have ambient listening in there during the surgery.

It's gonna capture everything and everything's gonna sort of work. I mean, that's, that's the dream. Right? But there really isn't an easy button because that's. Just one challenge. There's several challenges that, that have to be overcome.

Carla Haack: What's interesting is that it's also not that complicated.

It's not easy, but it's also not that complicated. It's a lot, and I don't wanna, I don't want this to sound disrespectful in any way, shape, or form, but it's, it's less stressful and less scary and a lot simpler than actually doing the surgery. Because what this is about is this is about making sure that you understand who needs to know what and when do they need to know it.

Bill Russell: right? That's beautiful.

, One of the things you said earlier on, I do want to go to use cases, if any that you could share I think are gonna be really powerful., One of the things you said earlier was, from a digital standpoint, you can change the tool, change it so that things are easier and you could change it so that they're harder.

And both of those things are in your toolkit. because there's some things you wanna make sure people don't do and there's some things you wanna make sure that they do. Talk about that a little bit.

Carla Haack: I could give you a couple examples. A denial example that I'll give you is the Z zero 0.00 code.

But it was also. The top diagnosis that was listed in the order when you opened up the order for, to associate the diagnoses. So we did a couple different things. One of the things that we did was, well, we worked with our primary care team to make the zoo code harder to find. But we also set up an a BN light type of workflow where we loaded the NCDs and LCDs from CMS into the system and we asked the system to fire an alert to a physician if they were ordering tests that were as and associating them to a diagnosis that was not covered, just saying, Hey, that's not gonna get paid.

By the time they see an alert once or twice, they're like, I'm not picking that one. That one's not, that one's gonna get me a popup and they pick another one. Right? Right. So you do the thing that makes it. Hard to do the wrong thing. You hide the zoo code and you let them know, and if they really insist on wanting to associate them, you make them type something in a field, right?

You just make it onerous and painful at the same time. You know, the example that I gave you about the debrief is an example of making it easy to do the right thing. And you don't even have to do anything different because what you're already doing in order to take really good care of that patient is gonna turn the rep cycle over downstream.

Integrated into the fabric of their coverage system. So if you're gonna be doing one of these novel procedures, you usually have to explain to the insurance company what you're gonna do because you're gonna have to use an unlisted code. And if you use an unlisted code and you don't tell the insurance company what you're trying to do, they're gonna deny you and they're gonna say you need to do a peer-to-peer.

because I have no idea what you're trying to precert here, so one of the things that we did was we built. The order set with the unlisted code. because one of the other things that was happening is that you would submit the unlisted code, you'd get denied, you'd get a peer-to-peer the, so then the next time the surgeon would say, well, the last time I submitted that unlisted code, I had to do a peer-to-peer.

So what we ended up doing was we built dot phrases. That would say things along the lines of there, despite, you know, based on my assessment of this patient would bene, would be best served by having a minimally invasive pancreatic debridement, despite the fact that there's ample evidence to support that this is a, a superior approach for this patient.

Based on the following factors, the code remains unlisted. We will perform this procedure, we will submit the unlisted code. However, the resources, competencies, risks involved in performing the unlisted code are similar to this comparable code. Therefore, we will submit the unlisted code and expect to be compensated in a way that is consistent or along the lines or in the ballpark of this comparable code.

Wow. Another example is when we found out that we had a lot of denials because we were precert the diagnostic maneuver without precert the therapeutic maneuver. There are situations like when you have lower extremity ischemia, where you have to do a diagnostic maneuver, an angiogram, so that you can understand what therapeutic intervention is most appropriate to perform on that patient.

This patient has lower extremity ischemia and they need intervention for the following reasons. When we go to the operating room, I'm going to perform an angiogram first so that I can un then decide what therapeutic intervention is most appropriate. Therefore I won't know until I get there. I'm going to precert all the possible codes and we will only bill you for what is actually performed.

And that went a long way towards helping us be able to precert all of the possible codes that we might perform. Make it easy for the physician to capture that in their documentation.

Bill Russell: I hope people are picking up on the fact that I said earlier, there's like a triangle, there's, there's it, there's finance, and then there's clinical you speak clinical very well, you speak.

from a reporting relationship where do you sit in the organization?

Carla Haack: I have a very diverse FTE, so I still have a, a, a toe on base in the department of surgery. But my FTE is mainly actually split between digital and finance. Currently digital is the majority shareholder by a small margin, and the rest of it goes up through finance.

Bill Russell: Finance. Interesting. We talked about a million dollars in 21 days in the or are there other examples that you can share?

Carla Haack: Oh man. So there are a lot of examples around plan mapping, cleaning up the plans, establishing so when we, when a patient would change their insurance, we would not pick up on that.

And, and based on how the RTE was running, because that was another thing that we found is that there were some groups that felt like they were empowered to reach out to the patient. And there were a lot of groups that did not. And that made it really hard to give our patients the information that they needed so that they could navigate our system effectively and, and with peace of mind, frankly, because that's the other part that I'm really passionate about, which is that when you could be taking great clinical care of a patient, but if you're not actually attending to the business side of their experience, they can suffer quite a bit and they can end up with big bills that you know.

So making it easy for people to reach out to a patient to say, Hey, can we talk about your insurance? Can we talk about how it is that you know. What did anything change? What's your plan? Finding our out of network patients was another really big issue that we actually had. So we had a ton of patients that were out of network that were getting care in our system, and that was bad for everybody.

There were a lot of those patients that needed to be cared for in our system because we offered services that weren't easily accessed in the community. So finding those patients. Grouping them into buckets of who's got something weird that they need to see us for versus who can be safely cared for by our colleagues at other organizations in the city, because they have great doctors and great teams that take great care of patients.

Bill Russell: And in network matters to me is the patient

If they're not in network, somebody's gotta write a letter of medical necessity. And those can take a really long time to write and they're extremely painful when they're time sensitive, you got a patient full, you gotta clinic full of patients or an OR full or day, and now you have to write this letter and it's gonna be really hard.

Find the time to do it. So leveraging AI to actually help write these letters of medical necessity so that we could more easily navigate securing a single case agreement for a patient or appealing a denial for that matter, was another game changer for us where it made it easier for us to do the right thing and follow the process that the insurance company has laid out for us to avail ourselves of those services.

Carla Haack: I will admit freely that I have been fairly obsessed with denials, um, since I've made this transition. And Epic told us point blank that our denial rate was the blemish on our implementation in as many words. When we went live, our denial rate was something obscene, like upwards of 20%. Wow. It was outrageous.

That was actually a real denial, by the way, which we fixed, because what would happen is that we would. If we had a patient, so many things that I wanna tell you. So hard to be cogent and coherent in trying to tell you all of these things that I'm so passionate about. Some of the denials that we had were deny what we called affectionately DINOs denials in name only, where we found that we were submitting charges for things that we knew were not going to be paid.

We should just stop doing that. There's no reason to do that. So there was a lot of that kind of cleanup happening. And there was also cleanup of self-inflicted wounds. We had this workflow for our third party sites. So when you submit a request for auth to some in certain insurance companies, you don't submit directly with them.

Which was then making that case drop out of the work queue, and then that number would populate the claim, and that would be an automatic denial because that number is not a real auth number. What ends up happening is that a few days later, somewhere between three and seven days after you submit your request for auth, the insurance company will come to that third party website, update that number with the actual auth number if they in fact approved the care.

So we didn't have a way. For us to go back and check that website unless somebody remembered that they had to go back and check the website, which wasn't necessarily the best use of our team members' time. So this is where we're talking about automation. Can we have a bot go ping the website and import that number when it changes?

Bill Russell: So there's an awful lot of this that is, organizational change management. It's meeting with teams. It's helping to understand where the process is fixed, educating.

Moving you know, changing behaviors, potentially changing some systems. And I mean, so there's a, you started this off by saying multidisciplinary I mean, your job every day is working with a lot of different departments in a lot of different teams, isn't it?

Carla Haack: Everybody. Everybody. I do not recognize Lane Dividers in my role.

That actually makes it so that the subspecialty is recognized. If you document the specialty before the subspecialty, then what ends up happening is that. Two different people in the same specialty, but different subspecialties may not be able to see and bill for the same patient as a new patient because the system perceives them as being in the same specialty and you keep two people in the same specialty, cannot both bill for a new patient visit for the same patient.

And so working with identity management to get our taxonomy set up properly so that the subspecialty is what is recognized, therefore allowing those two different subspecialists. To bill for a new patient visit with that patient the first time that they see them. So I'm partnered with identity management.

So. The job really consists of looking at the problem, trying to understand who you need to get together to solve the problem, getting all of those people in the same conversation, and then just making it easy for people to work together to fix stuff, which is incredibly satisfying.

Bill Russell: Yeah, I would imagine.

And then you say it in this other room and all of a sudden it doesn't, resonate would be the wrong word to say. It resonates in a way that, that is visceral, so you really have to, I mean, there's a, a lot of discernment. There's a lot of emotional intelligence to understand who you're in the room with and what is their motivation and, you know, and what their pain points are as and, and they're different.

Carla Haack: That is a very insightful comment. You are exactly right. So you don't go talk to the surgeons about NPSR in ebitda. That's not gonna get them excited, right? They're gonna glaze over if not have a visceral response. You talk to the surgeons about quality. You talk to the surgeons about efficacy. You talk to the surgeons about bankrupt patients.

Unpacking certain functional elements of the revenue cycle and of finance to physicians in a way that just helps them understand what you need to do about it in the way that takes up the least amount of your bandwidth possible. Right? How do I give my clinical colleagues a working understanding of how it is that their work gets credited or not, just so that they can maneuver more effectively?

because what ends up happening is that there's almost this hidden curriculum. In our clinical training that suggests, and I, again, I, I mean this very respectfully, but a lot of us carry around this impression that our job is to become very well versed on the latest and greatest.

Helpful for anybody. So I don't need my physicians to understand every nuance of how the revenue cycle works. I really need them to be focused on providing high quality clinical care in a sustainable manner. But they need to understand what words are gonna get them credit and what words aren't. And if they don't know, they need to know who to call.

So I do a lot of unpacking this with my clinical colleagues and I'm really grateful to have really strong relationships with the vast majority of my clinical colleagues, which helps tremendously because when I, they know they can call me number one. So now I get, Hey, they're telling me that my case for tomorrow isn't pre certed.

And I consider all of those in my purview and anything else that they can come up with. That I can do to make their lives easier and better because in order for me to make the decision to stop operating, which I ultimately made, I really needed to feel like I was not just serving. The population that I had been serving as a clinician, but a broader population, I needed to make sure that I was able to serve my community even more powerfully if I wasn't in the operating room.

And that means that if there is a problem and I can be helpful in solving it, I'm in.

Bill Russell: we're gonna close this out with three questions. Two of them are, are pretty straightforward. Denial rate was 20%. I assume that has gone down.

Carla Haack: We are looking at about 2% right now.

Would it matter to an organization if Rev Cycle were outsourced?

a lot of organizations have outsourced their rev cycle, and I'm wondering. If your job would be harder in, in those scenarios or it, you just have to have the right relationships and you just make it work.

Carla Haack: It's about the relationship. I really think it's about the relationship because you can have a brilliant rev cycle team, and if you don't have great relationships between your rev cycle team and all of the stakeholders that are involved, it's, it does, it's not gonna work at the same time, if you have a vendor.

That maybe hasn't been around for 20 years, but is, is engaged and willing to, to put in the work to make the relationship work. I can see that working well. So I really do think that it is about the relationship, but I will also say that there is a lot to be said for having a degree of organizational knowledge and wisdom around who do you have to go to?

And it's helpful to have that balance of. Organizational wisdom and knowledge and experience combined with a culture of continuous improvement, if you will.

Bill Russell: The final question, this is the exit question. I'm, I'm trying to figure out if I'm gonna do it as a two-parter. I think I am gonna do it as a two-parter.

Uh, the first part being so I'm listening to this and I'm going, oh my gosh, I've, we've, we've gotta, we've gotta hire for this role. We've gotta, we've gotta put this program in place. And that's the two questions really. It's like, what am I looking for when I'm going out to hire for this role? I've, I've decided I this, uh, what did we call it before?

Carla Haack: I will say that I have been very well served by having. A strong reputation amongst clinicians. So you, you want somebody who gets along with their colleagues. And I think that I have an advantage because I'm a surgeon and I get along with the surgeons really well, especially because I did emergency general surgery for such a long time that they called me for their issues.

And so they, I enjoyed that relationship. So I do think that having somebody who can. Have healthy relationships with your clinical team is an important characteristic. That doesn't necessarily mean that they have to be a surgeon. You just want somebody who's respected by the community. You also need somebody who has, and I, this is something that I learned the hard way, so I'm not gonna pretend like I always knew this, but who has enough emotional intelligence to bring a spirit of humble inquiry to conversations.

And finance is a little bit of a black box for some of these really smart people that. Admittedly may or may not have developed dysfunctional coping mechanisms to deal with the pressure and the chronically unmet basic needs of. Food and sleep and going to the bathroom and seeing your family. And so there's uh, that can actually make relating with people outside of that group more challenging.

And you can come into the revenue cycle in a conversation and say, well, what the heck is wrong with you people? And that's not going to go well. It's gonna go very differently than if you say, gosh. I don't actually know very much about what you do. Can you please help me understand your job and your pain points and how it is that we might partner to help everybody have a better day?

Bill Russell: Totally different. Yes. And, uh, good words of wisdom for the home too. Just

Carla Haack: indeed. Fair enough.

Bill Russell: Humble, humble inquiry goes a long way. It's like, it does. Why, why? What is making you upset? It's probably a better question than what's wrong with you. Anyway. I wanna thank you for your time. I wanna thank you for your work as well. And sharing it with the community here today. It's exciting and I'm gonna keep following up with you because I'm sure there's gonna be more use. Now that you're at 2% on, on denial rate?

What are you focused on next?

Carla Haack: So we're still chasing our denials. I think that we will always be chasing our denials and there will, you know, you are gonna fix one mechanism of denial and there's gonna be another one popup. You don't stay on.

Bill Russell: If you don't stay on it, it's gonna go from two to 5%.

I think that our clinical informatics team has done a really great job of thinking through. How can we make these workflows make sense from a clinical perspective and harnessing that and hooking it into the revenue cycle so that all of that work is really aligned, I think will also give us plenty of work to do going forward.

We've acquired a couple of hospitals, so bringing them onto Epic and, and all of those things are also large on the radar. We're talking about doing an ERP and I don't even know whether I was supposed to say that or not, but that's a humongous. Thing on everybody's radar right now. because what I've heard about implementing an ERP is that it's gonna make our epic implementation feel like we downloaded an app on our iPhone.

So

Carla Haack: absolutely. And the last thing that I would say that we're focusing on right now is that we've implemented strata.

We never had a tool like strata before in terms of time-based, activity-based costing. And there's a, a treasure trove to be uncovered with the information that we're getting from I

Bill Russell: love, I love how excited you get about. That's,

Carla Haack: it's really exciting because we've never been able to do this before. It's always been this sort of.

It was just excruciating to sometimes to get the data, and it required multiple people for a long time. Whereas now you can run these reports easily. So there's the bottom up thing of all the people texting me. And then there's the top down thing of the financial planning and analytics team looking at what Strata shows us and saying, well, we've got opportunities here, here and here.

Bill Russell: I will be quite frank with you to say, when I came into this industry and realized that we don't really have a good handle of our costs, I was sort of dumbfounded, like, because every other industry I've been in a lot of other industries, have a, I mean, they have a, a.

Laser focused on their costs, and so they're able to tell you it. It costs that to deliver that service or to do this thing, it costs X, Y, or Z. And I remember the first time I asked, well, well, how much does it cost to do that? They said, well, it depends.

Carla Haack: Exactly

Bill Russell: it was a learning curve for me to understand, why that's a challenge. But I think a lot of health systems now are looking at, Hey, do we really have a good idea of what it costs to have this here or to do this in this way?

Bill Russell: that's work.

And

Carla Haack: I would be So you're,

Bill Russell: you're not worried of working yourself out of a job, is what you're telling me. There's always opportunity.

Carla Haack: There is always opportunity and there's, I think that there, it's honestly now that I've been doing the work and finding so much satisfaction in doing it, to your point, it's kind of surprising that I'm the first one with this title.

And I certainly hope that more people will want to get involved. because it's a powerful way that you can make your healthcare system more effective in serving everybody in the community.

Bill Russell: Absolutely. Carla, thank you for your time.

Carla Haack: Thank you so much for having me. It's been an absolute pleasure and it is a privilege to be here.

Thank you so much.

Bill Russell: Thanks for listening to the 2 29 podcast. The best conversations don't end when the event does. They continue here with our community of healthcare leaders. Join us by subscribing at this week health.com/subscribe.