Abby Burns (00:16):
From Advisory Board, we are bringing you a Radio Advisory, your weekly download on how to untangle healthcare's most pressing challenges. I'm Abby Burns. In 2020, a team from the University of Iowa found that the cost to get a hip MRI in Iowa varied from $453 on the low end to more than 4,400 on the high end, 10 times difference. That phenomenon isn't new, and any quality leader in healthcare will tell you that the same variation exists when we look at care outcomes. Care variation reduction sounds like an operational solution to an operational problem, but I actually think that's an enormous undersell on both sides of the equation. CVR is an operational efficiency play, but it also ties back to both sides of the margin mission line that provider organizations are constantly walking.

(01:10):
Here's the thing, I think too often CVR work erroneously falls into the mental category of been there, done that when systems are looking for ways to improve operational efficiency, improve quality outcomes, reduce costs. Coding care pathways into the EMR doesn't do any good if no one uses them. So, let's talk about what good looks like. Last year in 2024, UNC Health relaunched a care variation reduction effort they initially started back in 2019. But let me clarify, they weren't relaunching because the first time had failed. They were relaunching because they'd achieved their initial roadmap. They'd addressed each of the 24 clinical areas on their priority list, and it was time to set their sights on the next wave.

(01:55):
Today, I'm sitting down with Cyndi Hall, Senior Advisor for Healthcare Plus Solutions Group, and former Executive Director of Care Redesign at UNC. And Dr. Larry Marks, Professor of Radiation Oncology, Executive Medical Director for Care Redesign, and Assistant Dean of Organizational Health and Analytics at UNC Health and School of Medicine. They're going to share the story of their last five years of care redesign work, what they did, how and why they did it, and the wisdom they've earned along the way about how to make this work. Cyndi, Dr. Marks, welcome to Radio Advisory.

Dr. Larry Marks (02:30):
Thank you. Pleasure being here.

Cyndi Hall (02:31):
Excited to be here this morning, Abby.

Abby Burns (02:34):
So, I want to start our conversation by grounding listeners in what is possible to achieve with the kind of work that we're going to be talking about today. So, my question to you all, to start us off, what is one example of something that UNC has achieved in the past five years through the care redesign work? Maybe a favorite example.

Cyndi Hall (02:57):
I will give you my favorite example, because I think it's something that everyone in medicine is probably doing the same thing or has done the same thing. It used to be, when Dr. Marks first learned to practice medicine, that if you gave someone a blood transfusion, you gave two units. That was the standard of care. That is no longer the standard of care, but we still see a lot of people doing that, or there's a lot of variation in practice patterns. At UNC, we brought in a ton of evidence, Dr. Marks can talk about the, what? The 34 studies, Dr. Marks?

Dr. Larry Marks (03:35):
30 something, I forget. 30 randomized prospective studies. Yeah, it's pretty amazing.

Cyndi Hall (03:40):
The evidence showed that's not necessary anymore. And so by changing our procedures, by educating our clinical staff, we decreased the percentage of time we give two units down to one unit up into the 80 percents. So we were really low in the 50 percents. We moved up into the 80 percents, and the real win is, in the first year alone, we gave 4,500 fewer units of blood. Blood is expensive. It's a rare, hard-to-get resource. This was a big win. It was millions of dollars for the system, and we've continued to do that. Dr. Marks, I think that the physicians, it wasn't even that challenging of a lift.

Dr. Larry Marks (04:25):
It's also risky. Every transfusion comes with the risk, so it was a win-win on all fronts. Well, the beauty of that example, Cyndi, is how easy it was to implement. It was an electronic thing that we changed in the medical record when you order the blood transfusion, it was actually less work for the providers. I think many of them didn't even realize it was happening when it happened.

Abby Burns (04:47):
The best kind of change.

Dr. Larry Marks (04:49):
So, I'm going to give two examples. The initiatives we did in diabetes made me very nervous. We did some things we had never done before, and we put in this new order set for taking care of insulin orders for diabetics, and it took several months for us to see a change in improvement. And those first couple of months when we had to revise the order set multiple times until we got it right, those were very nerve-racking times for me. And when that finally started to go down, I breathed a big sigh of relief. We had previously done many pathways in the operative space, but this was more of a medical thing, over a longer period of time, and it did take time and it has slowly continued to drop, and Cyndi, it continues to drop the hypoglycemic days in these patients.

(05:34):
But my favorite thing that happens is when I see a light bulb go off in one of my colleagues or a senior physician or a young physician who said he sees an improvement, and then says to me something like, "I get it now. I see the value." Because there's only so much that we can do in our group. It really needs to be buying into this culture and these principles. And the more that physicians and other leaders can understand these principles and apply them to all of their work, I think then we'll really see tremendous improvements.

Abby Burns (06:02):
What I really love about the examples too is they show the holistic range of impacts. It's not just financial. It's not just about improving clinician burden or patient outcomes. It really is across multiple domains.

Dr. Larry Marks (06:15):
Well, those things all go together. If you reduce variation and harmonize practices and make it easier for physicians to do the right thing, it's actually less work for them and the patients get better care. And if you do it right, you can improve operations and have financial benefits as well.

Abby Burns (06:34):
So, let's go back to the beginning to figure out how we got there.

Dr. Larry Marks (06:37):
Sure.

Abby Burns (06:38):
So, we're back in 2019, we're in a pre-pandemic world. UNC is chugging along as a state-funded academic health system, more than a dozen hospitals, thousands of physicians. At that time, if I have my math right, I think your oldest hospital had been operating for something like 130 years, but as a system, you'd been around for, I think, 67. What is it that, at that time, prompted UNC leadership to not only decide to take on care variation reduction, but invest in it heavily enough that they created a brand new office of Care Redesign?

Cyndi Hall (07:14):
In 2019, the system decided that they wanted to put together a huge transformation effort. Care redesign was born out of one of the three areas, which was transforming clinical excellence. And what ended up happening was myself and Dr. Cristy Page, who is now the Dean, she and I, and a gentleman named Tippu Khan, got together with 20 different leaders across the system and really started looking at care variation across multiple fronts. We looked at mortality, we looked at infections, we looked at cost as a proxy. We put together all this different data, including information from the advisory board. And after we did that, we found 20 areas where we thought there was high variation in care that would probably benefit from creating a standard.

Abby Burns (08:15):
So, Cyndi, it sounds like you were coming from a place of knowing that there was a lot of variation in the system, it was a problem that you targeted deliberately.

Cyndi Hall (08:23):
We assumed that there was. We just had to find where it was. All the evidence suggests everyone has variation. It's a known thing. And so, we came at it, we started looking for it and we of course found it. We found the usual suspects, and once we presented that to senior leaders across maybe I think I was in a room of a hundred people and we showed them what we had found and they said, "Yeah, we need to tackle this. We want to tackle this." And that's where we went forward. We showed the pro forma, we said, "This is a win. We can save money, we can improve patient care, we can improve quality. Let's go ahead and do this." And that's where the department was formed. Dr. Marks was the first employee. I was a second employee, and then we got started in December of 2019 with our first full team meeting, not knowing what was about to happen.

Abby Burns (09:18):
And I'm right that you all are the clinical and admin dyad over this work, right?

Dr. Larry Marks (09:23):
Correct.

Cyndi Hall (09:24):
Yes.

Abby Burns (09:24):
Was that intentional going in to have both roles on the team?

Dr. Larry Marks (09:28):
Yeah. UNC does this in multiple areas where we have a physician leader and an administrative leader. This is consistent with what we've done elsewhere.

Cyndi Hall (09:36):
It was actually one of my requests that I put in the business plan going out the door. I said, "I will not do this work, nor endorse this work, having just an administrative lead." This is so important to changing physician and clinical behavior that it has to have a physician voice and it has to have an administrative voice. And I think our partnership, Dr. Marks, I think it's been great. We've had a great time together.

Abby Burns (10:00):
I want to dig into how you got it started and how you got it off the ground, because I think coming with the brass tacks, coming with the, "Here's what we want to do and here's what we want to achieve with it," is the first step, and then you actually have to start. And so you identified all of these areas. How did you pick your original target list?

Cyndi Hall (10:20):
So, I made up a word, it's called septangulate, and it's like triangulate but with seven. And so, what we did is we had seven major data sources where we looked at every single DRG in the system, and we looked at the variation across these seven different fields of data, and we got the top 20, top 30, and we went through and then asked people, "Hey, do you think we can move the needle here?" Physician and change readiness was one of those considerations. At the early time when we looked at our OBGYN structure, we had some things we wanted to change. They just had gotten a new leader. That really wasn't the time. So, we put it on our roadmap, but we said, "This isn't going to be the first one out the gate because we want to give them the time to really get going and to get strong and to work together." And then when it was their time, they did a great job. But change readiness is important.

Abby Burns (11:19):
I think that is so important, Cyndi, and, frankly, that is the place where I think a lot of organizations underestimate the impact of internal readiness on the likelihood to succeed of any given effort, but care redesign, or care variation reduction being a huge one. What were the indicators for you, as you're looking at readiness in particular, that a service line was culturally ready to take on this work? And then I want to get into how you went about getting the buy-in.

Cyndi Hall (11:47):
The first thing we did, honestly, was ask. Just go to the senior leader and ask like, "Hey, how's your team doing right now? Are you stable? Do you have a structure in place? Are you ready? Do you think you can handle some change?" The second thing that we looked at is have they done any change recently? Have they done any group work? We did a lot with enhanced recovery after surgery, also known as ERAS. Best predictor of future performance is past performance. And so, if you have a history of being good change advocates, and that's something that we looked at.

Abby Burns (12:24):
I really appreciate that because sometimes we try to over-engineer or look for what's the empiric evidence, and sometimes it is the common sense or the body language of the folks in the room, or the straightforward things to look for.

Dr. Larry Marks (12:36):
The culture piece is critical, and particularly the leader buy-in, the leader of the units needs to want to do this. Members of the unit will follow typically or often follow the guidance of leader and the preferences of a leader. And so for me, it was always is there a strong leader who buys into these principles who's going to be an advocate? Because it's hard work, and it doesn't always pan out exactly as you plan, and you need to have that flexibility and understanding from the senior leaders that you're working with that this is a process, and it's an iterative process, and it's an ongoing process. It's not like you finish it and you're done, because the practice always changes, so you're always modifying a little bit the pathways and the workflows that you've defined. So for me, it's the culture of the unit, and particularly the leaders in those areas.

Abby Burns (13:22):
And you saw me vigorously nodding. I want to come back to the, "This is not a one and done, this is a culture piece." But I am curious, what work did you all do to get the teams that you were working with to the point where they were actively engaging and contributing? When you think about the service lines, the service line leaders, the teams, individual clinicians, how did you bring them along?

Cyndi Hall (13:47):
So, one of the things that we did in the early days is we said, "Hey, we're going to show you the best evidence that's known today to get this surgery done." And it's a lot like using a crowdsourcing app like Waze or Google Maps, et cetera. That software is going to tell you the best way to get from Raleigh, North Carolina to Chapel Hill, North Carolina. But it doesn't mean you have to go that way.

(14:16):
If you need to stop to get gas, if you want to pick up your dry cleaning, the application's not going to not allow you. And that's the same way with a clinical pathway. We're telling you the evidence best pathway to get from where the patient is to the really good outcome. Do you have to follow it? Absolutely not. But should you, because the evidence is there? Yeah, you would be wise to do it.

(14:39):
So, we set this up so our clinicians understand the concept, and then we take them through the steps of care in any pathway. And what we did then, is once we all agreed on the steps, we did something called a Delphi survey, where we send it out to every clinician, every nurse, every doctor, every anesthesiologist and physical therapist that might be part of this, and we let them vote on every single one of the elements of care.

Abby Burns (15:07):
So that's a pretty wide net for, "Are you going to take a right at the stop sign or are you going to go straight?"

Cyndi Hall (15:12):
Yes.

Abby Burns (15:13):
"Okay, we agree maybe that we're going to go straight. What's the next step? Let's all vote on that." That's the method you're talking about?

Cyndi Hall (15:20):
We are. And what's interesting about this method is, we didn't vote and say, "Hey, what do you prefer? Left or right?" We say, "Hey, the evidence says that making a right-hand turn's faster and safer than left." We have all chosen to say right, but we're going to give you a chance to vote. And if you vote, "No, I disagree." You can't just say, "No, I don't want to." You have to say, "No, because of study X, Y and Z by this group." You have to have a reason for your no.

Dr. Larry Marks (15:49):
So, what we did was, we would come up with this pathway that Cyndi has described. We would then send it to the stakeholders who would be using this pathway, get their feedback. And what she was describing where they would get to vote on each of the elements in this pathway, we would use this thing called the Delphi method where based on the feedback we got, I forget the threshold, I think we needed agreement that 85% or something like that-

Abby Burns (16:16):
Which is pretty high.

Dr. Larry Marks (16:17):
Level. Pretty high. So we would iterate. There'd be might have 20 or 30 items the first time through, and then five come back where agreement is only 50 or 60%. You look through that, you go back to the team and you say, "Hey, this is the feedback we got from the system, from the Delphi method. What do you think? Should we make changes?" And we would tweak it a little bit and we would then send it out for a second survey. And then of those five elements where there was disagreement, now all of a sudden three of the five, you're at the 80% and you're okay with it. The other two, you do the best you can and you try to get that as high as we could.

(16:51):
But that process, I think that people appreciate being asked. And physicians and nurses and all healthcare professionals, they're proud of their work. They're smart people went to school a long time, and we want to be sensitive to, I don't want to be seen as Larry Marks and Cyndi Hall telling physicians how to do their job. I don't know how to do their work. So, we want to get their feedback, and we want to build pathways that they're comfortable with because otherwise they won't use them. We could build a pathway, if they don't use it, what was the use?

Abby Burns (17:23):
Hugely important point.

Dr. Larry Marks (17:25):
We need to be realistic to build tools that people are going to want to use. So we want them to be medically acceptable to them, and we want them to be good. We want them to save them time, save them clicks, make them feel more comfortable with the care that they're providing. So it's a delicate balance. Analogy Cyndi used about the Waze, the driving app is a really good one. I think physicians in particular get very upset if you say, "Don't tell me how to take care of my patient." I'm not. We're not doing that. We're giving you a recommendation and you could modify if you want, but does this recommendation work for most patients? And if the answer is yes, we hardwire that into the electronic medical record. We make it easy to do the right thing.

Abby Burns (18:04):
You're not saying, "Get in this self-driving car. It's choosing how you're getting to the grocery store"? The clinicians are still driving their own car. They can follow Google Maps or not, but they know the route.

Dr. Larry Marks (18:17):
Yeah. I like to make the analogy of for the stuff we can agree on that's going to work for most patients, let's hardwire that. Let's make it easy to do that. Let's make that the cookbook, if you would, or the cookie cutter. You can still modify it, but if you don't modify that and that cookie cutter works for most of the elements, you've now freed up their brain and they now have the mental bandwidth to think about the particular elements that really do need to be modified.

Abby Burns (18:43):
The two things that jump out to me for that, one is, the emphasis that you all have put on the clinical evidence being so integral to the pathways that you ultimately decide on, even from when you get pushback of, "I don't agree with this pathway because here's what the evidence said." This feels so important in a world where clinical evidence is, what, doubling every 73 days? It is impossible for any clinician to keep up with all of the latest studies. I think that estimate is actually even outdated as of now, so it's probably even shorter. The other thing that I want to double click on though is you said that your threshold was around 85%.

Dr. Larry Marks (19:22):
Maybe it was lower. Cyndi, remind me, what was it? Do you remember?

Cyndi Hall (19:26):
It's interesting. We didn't have a hard and fast rule.

Abby Burns (19:29):
Okay.

Cyndi Hall (19:29):
We loved to see 80 to 85%, but that wasn't always possible because some pathways, especially medical pathways, not surgical, have much more patient variation, and so we might set a lower threshold for that. One of the things that we read in the Advisory Board early on was something called the innovation zone. And what that is between 60 to 80% of all variation work, that's your threshold. You want to hit for 70, with 60 to 80 being your range.

(20:01):
The reason for that is really important. We want our clinicians to have the ability to change because they might find a better way. We want them to innovate, to try things that are different for different patients, and we've done that with our pathways at UNC. We've gone back and said, "Oh look, when this one antibiotic was used instead of this one, we're seeing better outcomes. Let's go ahead and change the pathway now that we have that new information." I'm looking at my tip sheet right here in front of me. With that blood patient management group, we had a cohort of 32,000 patients.

Abby Burns (20:43):
That received the single transfusion versus the double transfusion?

Cyndi Hall (20:48):
That was getting a transfusion of any kind during this time period. So, for those people who aren't familiar with blood transfusion, this is over a long period of time. This is not a year's worth of transfusions. But we got really good information from that time period and we were allowed to tweak things. We're allowed to change things for our enhanced recovery after surgery. We learned that when a woman has a C-section, one of the greatest indicators of how she's going to do afterwards is how well she maintained normalthermia or a normal body temperature during her procedure. When she maintains normalthermia, she's much less likely to get an infection. That's information we didn't have going into the pathway, but then we prioritized it after we started seeing the evidence come in.

Abby Burns (21:38):
This innovation zone that you're describing feels increasingly important as clinical conditions meld more. I'm thinking service lines, historically, maybe have been pretty siloed. If you're dealing with heart failure, maybe you're dealing with heart failure. If you're dealing with a GI issue, maybe you're dealing with a GI issue. Increasingly when we look at things like managing obesity or other chronic diseases, they are interdisciplinary treatment pathways. It feels really important to have the tools available to clinicians to be able to navigate crossing paths or crossing service lines accordingly.

Cyndi Hall (22:16):
Yes. Abby, you bring up a really good point. And the way healthcare gets paid is by specific disease state. Unless you're looking at value-based care, we don't get paid across the continuum. The reason why that's important is because that's then how we practice medicine. You tend to practice in the silos in which your disease state works. We have found that as we've gotten better and better at pathways, we are able to move out of what we call these vertical pathways, which is a specific surgery, a specific disease state like pneumonia.

(22:55):
And we've been able to start doing cross-continuum work, also known as horizontal pathways, and that's been something that we've done, Dr. Marks, I would say, just really in the last year or two where we're starting to get good enough that we can look at a patient's care from when they present to the primary care doctor with knee pain, and what are the right steps before they decide that they need to have a total knee replacement because there's a lot of variation.

(23:23):
I think we looked at the different number of ways that a patient experienced knee pain, and there were 14 different unique paths between physical therapy, the surgeon office, dietitian, all this crazy stuff, different patterns where people would follow. And there really is a really good way to do it, but it's just hard to navigate.

Dr. Larry Marks (23:46):
The part of that that I remember most is the imaging that the physicians and other providers order upstream before they send them to the orthopedist is almost always the wrong imaging that the orthopedist needs to reorder, therefore images when they see the patient.

Cyndi Hall (24:02):
It was only right 15% of the time. 15% of the time they ordered the right images. The rest of the time the surgeon had to redo it.

Abby Burns (24:09):
Which is getting into almost waste. The wasting system.

Dr. Larry Marks (24:13):
Well, remember the physician's doing this or the provider's doing this, they're not doing this on purpose, but they mean well.

Abby Burns (24:18):
Absolutely.

Dr. Larry Marks (24:20):
When the truth is, when you go into the electronic medical record to order a test, it's not so simple. There's multiple to choose from. It's very confusing. So, you could see why the incorrect one was ordered. So, they're well-meaning, but having we now have these pathways for the preoperative evaluation of patients who just have knee pain, who you would considering sending them, if you order the imaging on the pathway, the correct imaging is ordered. So, therefore, when they subsequently see the surgeons, they have the right imaging there, so the surgeon doesn't have to reorder things. So, it's less expensive, saves the patient time, saves the doctor time, and it's better care. Saves society money.

Cyndi Hall (24:57):
One of my favorite stories, Abby, is there was a physician, Dr. Tony Rodriguez, and he's a family care physician. And he had a female patient come in presenting with knee pain, and she just wanted an injection. That's all she wanted. She's like, "Just give me an injection in my knees. That's what I need." You talk to friends, you Google things. He said, "Let me show you this pathway." And he literally pulled up the pathway in Epic, in our electronic health record and he showed her.

(25:25):
He's like, "Look, these are all the different things we can try before we go to an injection, which has risks. It's an invasive thing. They don't always go perfectly well." She looked at it, and they entered into shared decision-making. She's like, "You know what? You're right, I haven't tried this. I haven't tried a walking program. I haven't met with a dietician to lose weight." And they came to an agreement together on what her care plan should be, and moved her forward using evidence, using shared decision-making. It was the perfect example of what we want healthcare to be, to use appropriate resources at the appropriate time for that patient. It was a beautiful moment.

Abby Burns (26:06):
The patient experience or patient trust in their provider, that transparency and that confidence of the care that I'm receiving is so clearly evidence based. I think that is a standout example.

Dr. Larry Marks (26:19):
I want to add something to that concept. So, we love data, we love to make the pathways based on the evidence of what the best care is. But what do you do when there is no data?

Abby Burns (26:32):
Important question.

Dr. Larry Marks (26:33):
It's a conflict I have with some colleagues and vigorously debated issue. So some people will say, "Gee, there's no data, therefore I get to do whatever I want because I have an opinion." So now what do you do with that? And it's a challenge because you out of three providers, and one wants to give aspirin, one wants to give Tylenol, and one wants to give Motrin. And they may all be reasonable. And those choices though have upstream and downstream consequences to the pharmacist, to the nurses, and to other members of the care team.

Abby Burns (27:04):
Can you spell out the impact to the nurses in particular?

Dr. Larry Marks (27:07):
I'll give you the insulin example is a better one. You have the same nurse taking care of two patients with diabetes, and the orders for insulin are slightly different for patient number one and patient number two. Both orders are reasonable, the physicians ordered the right thing, but they're slightly different. This one nurse taking care of these two patients has to remember that they're slightly different. It just leads to unnecessary mental bandwidth that the nurse has to expend on that, and we're short of nurses and nurses are overworking. So, this concept of there is value in reducing variation even if there's not great evidence that one is better than the other. All right-

Cyndi Hall (27:48):
We call that practice-based medicine. You have evidence-based medicine and when you have practice-based medicine. And to Dr. Mark's point, which is beautiful, clinicians and patients benefit from the reduction in variation regardless of what we use usually. Sometimes it's been wrong, but usually our data, we can tweak it soon enough that we can get it right back.

Dr. Larry Marks (28:11):
The thing that's interesting to me is they probably benefit in ways that we can't even appreciate. Just by reducing the mental burden on that nurse, that nurse may therefore be more likely to suspect sepsis earlier in the patient, three beds over.

Abby Burns (28:26):
Or even just have a better bedside manner.

Dr. Larry Marks (28:29):
Yeah, sure. Because they're less frazzled. But it could be on other patients. So, harmonizing and making the care easier for patient number one may improve the outcome of patient number two because we freed up the time for the providers to think about the stuff that they need to think about. So we reduce the burden of the stuff that they don't need to think about.

Cyndi Hall (28:51):
And I want to tell you, Abby, we actually track that a little bit. We look at our patients that are on pathway compared to our patients off pathway, and we actually see, because we pull in our Press Ganey scores, we see that our patients on pathway have a better patient experience than our patients off of pathway, and that is really helpful to see. The likelihood to recommend is just higher, generally speaking, when a patient's on a pathway, and that's probably exactly because of all the things we just spoke about, the nurses know what they're doing, the physical therapists know what they're doing, the doctor is being more consistent. So, everyone is getting this known quantity, but we still allow for that variation of precision medicine where the N is one. We're going to do standard of care, except in those places where you are different and unique and we're going to meet you right there.

Abby Burns (29:46):
And being able to be a little bit more precise about when is the situation truly unique versus when are they appropriate for the pathway. One very practical question I have, tying the patient experience scores back to on pathway versus off pathway, is that hard to do? Can every system do that?

Cyndi Hall (30:06):
Yes and no. If you warehouse your own patient experience data, meaning you download it and you put it in and it gets tied in, then yes you can. If you're not currently doing that, you would have to move to that as a data point. But yes, everyone theoretically can do it if you download the data.

Abby Burns (30:24):
Yep. I'm curious, as you're approaching the care redesign work or as you were approaching it in 2019, how did you define success for this work?

Cyndi Hall (30:34):
Our original success measures were going to be both administrative as well as clinical. We wanted to do things like reduce readmissions, reduce the number of times that people went to the ED, also known as a bounce back, and reduce length of stay, how long the patients were staying, which sometimes shows evidence of inconsistent care that we just don't have our act together and they end up staying more days than is normal. But then we also looked at clinical outcomes. How many hypoglycemic events are patients having? How many infections are people getting? What's the mortality rate for certain procedures? So we looked at all of those different things, and those would be what I would consider the clinical wins.

Dr. Larry Marks (31:19):
And for each of the pathways, we would have particular predefined measures of what was going to be monitored. So, for example, the one that comes to mind for me is the high-sensitivity troponin in the ED, immersion department. There was a new test, a high sensitivity troponin, and there was some lack of clarity amongst the providers. What's the optimal way to interpret this data? So it hard-coded for the providers a standard way to interpret this data. When we turn this on at the various hospitals, the admission rate for patients with chest pain dropped 7%, absolute percentage points, essentially immediately. And so that was good.

(31:58):
So, we saw that dramatic reduction in admissions. You could say, "Gee, we avoided those admissions." Yes and no. We have to always have a balancing measure. Are we causing harm? So, in this case, just because we didn't admit them to the hospital, that isn't the win. We have to be, "We didn't admit them to the hospital, they went home, and they had to be okay. If I sent them home and they die at home, that didn't help." So, we would follow, in that setting, a thing called the major adverse cardiac event rate over, I think, it's 30 days. So, we very carefully, we saw the admission rate went down, but then we were very careful to make sure that those patients were still having good outcomes after they went home, and we indeed confirmed that that was okay, they were doing okay.

(32:38):
So, for most measures, the direct measure of is clinical care, is the outcome better? And then, it's a balancing measure to make sure that we're not doing something that has some unintended negative consequences.

Abby Burns (34:37):
Earlier on, Dr. Marks, you said this is not a one and done type of thing. This is an ongoing way to practice. I know that last year in 2024, you all started essentially the 2.0 version of the care redesign work. Why now?

Dr. Larry Marks (34:57):
When I said that earlier, what I was meaning was the clinical care is always evolving and therefore the pathways always need to be tweaked over time to accommodate the changes in practice. The 2.0 you're referring to now is the approach to building these pathways, as we described before, is a very labor-intensive manner, taking many, many months to get broad input from many people to build these pathways. And we've built these pathways successfully. Over time, there's a desire for more and more pathways. So there's a new product called AgileMD, it's a piece of software that sits on top of the electronic medical record, and it makes it easier to implement pathways and easier to modify pathways once you've built them.

Cyndi Hall (35:42):
The perfect example of that I will tell you is when Hurricane Helene hit in the western part of North Carolina, it took out one of the IV solution making factories, so all of us had less solutions available. Within 6.5 hours-

Abby Burns (36:02):
Hours?

Cyndi Hall (36:03):
Hours of recognizing and saying, "We can create a pathway for when it's appropriate and you have to give the IV solution versus when you can hydrate your patients in a different way." We set that pathway up. We had all of our experts in the same room on the same phone call. They reviewed the pathway, and we got out the door into the hands of 18 hospitals, but especially our ones that were right there, in the same day.

Dr. Larry Marks (36:33):
In the same day.

Cyndi Hall (36:34):
That was amazing. We could not have done that three years ago. We could not have done that without the support of AgileMD, and we can track the outcomes and we can track what has happened, and that was remarkable.

Dr. Larry Marks (36:45):
It was very early in our days of knowing how to use Agile, and our colleague, Claire Niece, built that pathway, if I remember, in one day.

Cyndi Hall (36:52):
She did, yes. Fantastic RN. She's one of our directors in care redesign.

Abby Burns (36:56):
That is amazing. Not to do the play on words, but that is amazing agility in response to an actively unfolding disaster to be able to roll out that type of support to your clinicians.

Cyndi Hall (37:10):
Just so everyone hears this loud and clear, especially my colleagues at UNC Health, all pathways will not be that fast. But we really have sped up. In the beginning, it would take us six to 12 months to get a good pathway up, and now we're seeing a up in six weeks, four weeks. Part of that, and I know, Abby, we had talked about this in a previous conversation. We have the trust now. In 2019 and 2020, we were new. We didn't have the trust of our clinicians. We didn't have the numbers to support this fast-paced change in pathways. As we built our culture of trust, as the physicians and the clinicians came to believe in care redesign and the importance of reducing variation, that allows us to go faster now. You can build pathways without trust, but it's going to be a slog.

Abby Burns (38:05):
To close out our time, I want to hand the microphone over to you all. What guidance do you have for other leaders, other organizations that are considering or maybe should be considering taking on care redesign work?

Cyndi Hall (38:23):
I think that what leaders of a hospital need to do if they really want to do this care variation reduction work, is they need to dedicate themselves to it. The risk is to make it the project of the year or to make it a pilot. If you really want to do this work, your senior leaders and your physicians especially need to commit to CVR, to care variation reduction. We set up a department, we set up a dyad, we started building the trust and using evidence. I think that's the right way to do it. I think if you just try to have two or three small little areas, just do a few things here and there. People that are resistant to change, not the early adopters, but the other end of that adoption curve, they're just going to wait it out. They've been there for 10 years, they're going to be there another 10 years. If I wait long enough, I don't have to do the change, and you have to get away from that mindset.

Dr. Larry Marks (39:22):
And the leaders also need to allocate the time from their people. The physicians need to be freed up of the time. The nurses, all the stakeholders need to have time to go to meetings to get involved in this work. If it's an add-on, when people are tired, they just may come to resent this sort of work. It's an important component of what everybody should be involved in.

Cyndi Hall (39:47):
If we have time for one more analogy, I will tell you that it's one thing to create the pathway, but that's not enough. You actually have to look at adherence. The story I always tell is I can get all of my friends and families to join Weight Watchers. They can even go to the meeting and download the app. That does not mean they're losing weight. With care variation reduction work, you not only have to set up the pathway, but you have to look at adherence. In those enhanced recovery after surgeries, are they actually giving the multimodal analgesics? Are they actually walking the patient right away?

Abby Burns (40:25):
It's concept into practice. It's not just concept?

Cyndi Hall (40:28):
Yeah, and because a lot of people think, "Oh, I'm doing pathways. I'm doing care variation reduction work." And they've set up the pathway, but they're not monitoring the results. They don't actually know if they're adhering to the pathway. That's where you really get good, is when you start having those conversations with your clinicians and the trust is built.

Dr. Larry Marks (40:49):
On some regard, it's on us if we're going to build a pathway to make that pathway easy to use so that the providers want to use it. Providers, we're human, and people will make decisions and for a variety of reasons, but we need to make it easy to do the right thing.

Abby Burns (41:08):
Cyndi, Dr. Marks, thank you for coming on Radio Advisory.

Dr. Larry Marks (41:13):
Our pleasure. Thanks for having us.

Cyndi Hall (41:15):
Had a great time.

Abby Burns (41:22):
What I heard from Cyndi and Dr. Marks is resist the urge to think of CVR as a, been there, done that sort of thing. Resist the urge to make it the project of the year. To see the kinds of clinical, financial, operational, strategic gains that they were talking about, they really leaned into building the infrastructure and the culture of reducing clinical variation. That's what let them move quickly, and that's when you can start to take things to the next level and address variation across service lines even in the ambulatory space.

(41:56):
Reducing clinical variation is something Advisory Board is actively researching in 2025. If CVR has been on your organization's docket and you want to share your work, reach out to us at podcasts@advisory.com with the subject line, "Sharing Our CVR Work," to get in touch with our research team. And remember, as always, we're here to help.

(42:18):
If you like Radio Advisory, please share it with your networks, subscribe wherever you get your podcasts and leave a rating and a review. Radio Advisory is a production of Advisory Board. This episode was produced by me, Abby Burns, as well as Rae Woods, Chloe Bakst, and Atticus Raasch. The episode was edited by Katy Anderson, with technical support provided by Dan Tayag, Chris Phelps and Joe Shrum. Additional support was provided by Leanne Elston and Erin Collins. We'll see you next week.