S4E3: ROI Via Centralized Utilization Review (ft. Josh Moskovitz, NYC H+H)

Jordan Cooper   0:58
Mhm.

Joshua Moskovitz   1:14
And so how can we be better, smarter and more effective and efficient with it? And one of the things we noticed with, you know, with ED utilization review, emergency department utilization review is that between the 11 acute care facilities we have, everyone was sort of doing it a little bit differently. Some were doing 24/7, some weren't.Some had the nurses doing multiple tasks like care management tasks with the EDUR tasks and it really was setting up a system where it wasn't as efficient and effective as it could be. And so we said what can we do that's different? And we looked at it and really the skill set and the roles of EDUR while having a lot of overlap with CM is distinctly different and to do both at the same time.Is very distracting and so that let's carve out the EDUR piece and let's systematize and centralize it, especially because we are live on Epic now for for many years and we have a system we can communicate with people electronically. Why not centralize that process, build it up, develop it, invest in it and make the best quality candidates working on it as possible to be most efficient?Infected in that universe.

Jordan Cooper   2:13
And just for our listeners benefit, can you outline some of the key stakeholders involved in the decision to move ahead with this process?

Joshua Moskovitz   2:23
Yeah, it was a system wide conversation. It was system wide between nursing leadership, physician leadership, care management leadership, finance. It's really EDUR touches every point of the healthcare journey on an inpatient side and so really quite a big global strategy. It was not a short or quick undertaking, it was.

Jordan Cooper   2:35
Mhm.

Joshua Moskovitz   2:40
Months to years in the making, we officially went live with a centralized EDUR team as a one year. Actually, our one year anniversary was this month and so now live and active for one year. But the process to build up the tools and the resources you can get there were, you know, many months to years prior to the launch of it.

Jordan Cooper   2:48
Mm-hmm.So I'd like, you know, a lot of times at Healthy Data podcast, our listeners are are listening to this episode and we'd like to make the stories real to them. Can you walk us through an example of, I guess, how the new centralized EDUR is impacting some sort of end user's life, how?Data is different and what the impact of this change has been.

Joshua Moskovitz   3:21
Yeah. So the impact is, is really great. So #1 is we're doing the reviews in in real time as possible, right. We're really hitting the reviews within an hour or two of being admitted to the facility. So it's really allowing real-time conversations between clinicians and the nurses doing their reviews. And so in the world of billing and reimbursements, you know, unfortunately the medical record is still that Ave.

Jordan Cooper   3:27
Hmm.

Joshua Moskovitz   3:41
Conduit. And if it's not in the medical record in a form of medical decision-making or a form of a diagnosis, you're not going to be able to get those costs provided. And so these nurses are reaching out to physicians in real time and ask them to help upgrade the documentation, upgrade the quality of their level of care or upgrade the diagnosis.

Jordan Cooper   3:49
Mhm.

Joshua Moskovitz   3:57
Example, a physician admits a patient for leg pain. That's really not a great diagnosis. In real time, they can look into the chart and recognize the MDM is pretty much in there talking about this is a gangrenous foot that's going to go to the OR and it's really an acute emergency. And so by communicating with the physician in real time, we can show data where the.Diagnosis are increasing, the level of severity of illness is increasing and being more reflective of actually the care providing.

Jordan Cooper   4:24
I appreciate that. I think that probably allows for greater revenue capture for the organization. Is that right?

Joshua Moskovitz   4:30
Absolutely. And also denial mitigation as a result. So we're watching our denial rate go down, revenue go up and also trying to improve the the severity of illness of our patients and the CMI of our patients by more accurately capturing the information on the medical record.

Jordan Cooper   4:45I got it. And so in terms of and so it improves, is it primarily improving documentation that leads to those outcomes?

Joshua Moskovitz   4:54
It's a combination. So #1 is the documentation on the front of the physician, right? So the ED physician is that first contact point to the process of where denial happens. It also goes through the inpatient care stay, but the EDUR nurse manager also is responsible for taking that initial documentation and packaging of arrival letter and saying to the insurance company explaining why they're.

Jordan Cooper   5:02
Mhm.

Joshua Moskovitz   5:14
Being admitted to the hospital. And so you know, if the physician documentation is not written in a manner that's reflective, for example, syncope as a case that gets denied a lot, this is syncope and they send it another way. The care manager's massaging out the details and saying, wait a minute, you're describing heart failure in your, describing a kidney injury, describing a head injury in there. I could see by your thought process by what you're ordering that's.You're thinking, but it's not reflecting. Let's do that. And the care managers getting better at writing that letter, applying the criteria that a lot of the insurers use to judge an admission and then sending it off to the insurer. And so that only not only improves our initial upfront payment, but also our initial upfront length of stay needs for the patient. You put a week diagnosis like syncope, you get a day out of it.

Jordan Cooper   5:35
Mhm.

Joshua Moskovitz   5:54
You put a cute heart failure on chronic heart failure, acute kidney injury and a head injury. You're talking about a couple more days of hospital stay automatically approved up front.

Jordan Cooper   6:02
So it sounds like there's a lot of back and forth between the payer arm and the clinical arm when it comes to utilization review. I'm wondering if any agentic Gen. AI tools are in the roadmap for being leveraged in order to facilitate this back and forth.Work.

Joshua Moskovitz   6:22
Yeah, we're actually looking into some A I stuff, not necessarily for the back and forth piece of it, but actually reviewing the documentation to improve the quality. Like are there better diagnosises that can be captured? Can we be more accurate and more specific in the ICD 10 codes when you're on there? You know, you talk about in the past issue this CK day, you talk about the API today, but you don't put the two words together and you choose the wrong diagnosis.

Jordan Cooper   6:29
Mhm.

Joshua Moskovitz   6:42
This is on the IC10 list that's going to lower your index. So from a data perspective, we're using this to improve the CMI, improve the data capture, improve the quantity diagnosed in the chart to be more accurate or popular with the care we're providing.

Jordan Cooper   6:44
Mhm.Mhm.So Josh, I think many of our listeners might be interested in a few things. The first of which would be what are some of the data challenges of going through a years long process of centralizing this EDUR overhaul?

Joshua Moskovitz   7:09
Yeah, so that's a great question. That's bad data makes for bad output, right? We know this garbage in, garbage out. And so prior to launching this, we spent a good year fixing the quality of the data capture on the physician and the user. So for example, we created a standard system-wide note that helps with medical decision-making when you're admitting a patient.

Jordan Cooper   7:14
Yeah.Yeah.

Joshua Moskovitz   7:29
And we automate it so that if you choose a disposition admission and you open this note, it automatically pre-populates with the questions that we know you're thinking about, we know you need to answer. Why are you admitting this patient? Because the labs are abnormal, the imaging is abnormal, they're at high risk. So it captures all those data points.That are relevant to the admission and put it on a piece of paper right there. That will then automatically pull in to the Ed care manager's note as they're writing the note. So a lot of data is being pulled in all the Admiral vitals, the Admiral labs are being pulled in automatically now to make their time effective and efficient.

Jordan Cooper   7:58
Mhm.

Joshua Moskovitz   8:01
On the next point, we've now standardized the inpatient documentation across our health system. So all inpatient medical documentation is problem-based charting now and it pulls in all the data on the chart and allows you to update it in real time. So when the physicians, especially the residents, are writing notes.They can update the course throughout the entire hospital state to make their time more effective and efficient and more reflective care.

Jordan Cooper   8:23
OK. So it sounds like there were a lot of efforts before implementing to fix the quality of the data and the big path to to do that was to standardize documentation. Is that accurate?

Joshua Moskovitz   8:38
That a huge lift of ours, a huge foundational improvement to get to the point we're at, yes.

Jordan Cooper   8:42
OK. And then from there you were able to automate some of this note documentation to facilitate greater time to admission and greater throughput and efficiency, which you mentioned the very beginning is all pointing back to the need to make.More happened with less as a public safety net health care delivery system, correct?

Joshua Moskovitz   9:01
Absolutely. And we're finding the quantity of reviews. So when we measure the Edu activity, we're measuring first the quality of their views, making sure their reviews are reaching the quality expectations we have. We have 5 standard points we look through as far as the quality, are these in the right data, the right care sets, the right etcetera. And we look through the quality and watch that improve over time and then the second stage is to improve the.

Jordan Cooper   9:08
Mhm.

Joshua Moskovitz   9:20
The output of there, how many reviews are they able to do per hour? Are they able to, you know, do a certain amount of cases per day? How many touch points are they having with the physicians? And so we're measuring that and we're watching their quality and quantity both increase at the same time.

Jordan Cooper   9:33
Well, I think many of our listeners would love to implement something like that that increases the quality and the revenue for the organization while while improving patient outcomes. I think you also work on a process of centralizing writing appeal letters for your health system.Can you delve into what that process looked like?

Joshua Moskovitz   9:55
Yeah, that's still in its early stages right now. We're at the point where we're about to pilot a lot of that requiring to kind of in the background organize flow and throughput and outcomes in Epic to allow us to streamline work through processes. And so we're looking to take the appeals writing that was being done locally the sites in various different formats, a lot of outside Epic formats on Excel or.Or Word documents or an e-mail really trying to bring it in-house so everything is searchable, quantifiable and reviewable. And so in this process creating work queues and work lists that will feed the information along the the conveyor belt of data that needs to go through. But then also as the physicians are writing their appeals quantifying the the defensibility, this is a high.Defensible case. This is a less defensible case. So then we can then kind of go back and measure the data and say where are our problems? Where are cases that are clearly appropriately being admitted and treated for but not being paid by the insurers? Are these?Inappropriate denials are these, you know, these regious denials should be going after the insurers in these cases where they're not providing the reimbursement for the care provided versus these are cases that are less weak, less strong. It's just a part of our patient demographics and population and their healthcare needs and it just became more complicated and part of our mission-driven care.

Jordan Cooper   11:09
What are the data sources and data destinations for this centralized process of writing appeal letters?

Joshua Moskovitz   11:17
So when the appeals are received at a local facility, they're injected into the system and then fed into a work queue, which is then routed to a global group of physicians. What we're working on right now is to have a global group of physicians ready to write these appeals, which will then kind of route it back to local facility to finish off the process of.Sending back to the payer for appeals, that's kind of the 30,000 foot view over there process of the position right appeal and they're looking through the charts. They are as they're processing it, they're giving their opinions on the level of quality of the case and you know whether it meets criteria, doesn't meet criteria, whether it follows clinical guidance.

Jordan Cooper   11:40
Mhm.

Joshua Moskovitz   11:55
Versus one of these more specific systems of of pair guidelines.

Jordan Cooper   11:55
Yeah.A lot of times I ask guests on the show to quantify the costs of implementing new programs, and the cost could be financial or it could be cultural. It could be getting buy-in. And the idea is to help evaluate the return on investment of of such interventions, however.It seems like both of these centralized process of processes of centralizing Ed utilization review and writing appeal letters is increasing revenue with the accurate charge decreasing denials.Of claims and then leading to greater capturing of claims that have been submitted and denied and identifying which are egregious and may be able to be reversed. So it sounds like the financial case for both of these interventions is is conspicuous, is that right?

Joshua Moskovitz   12:58
Yeah, I would say the one thing that you didn't capture on there is also the free of resources, right. So we don't capture very well is when you get a denial and how much time, staff and energy is wasted on chasing it, right. So it comes in, the local person opens up the envelope, they process it, they they file it, they.

Jordan Cooper   13:09
Mhm.

Joshua Moskovitz   13:14
Notify someone, someone as a researcher, look into it, decide they're gonna act on it. If they do act on it, you gotta write it up. And the whole process of doing this, this, this whole workflow here. And it's it's hard to quantify how many touch points I have. It depends on where you're talking about. Some institutions, I think a denial will forge written off makes its way all the way to CF O's desk to say, you know what, we're not gonna chase this anymore. Write it off.

Jordan Cooper   13:33
Mhm.

Joshua Moskovitz   13:33
And so how much is that person's time worth, right? So we think about all the touch points along the way. Every denial saved also frees up staff to spend their time on more valuable activities than chasing denials.

Jordan Cooper   13:45
So are there any other avenue, any other areas of the organization where centralization is going to benefit the organization? Or an alternate question would be, are there any downsides to centralizing? Because it sounds like there's a lot of value here.And.

Joshua Moskovitz   14:04
Yeah, I mean, it's a big culture change and could culture shift, right? So one of the hardest parts is 11 facilities do things 11 different ways. And so how do you get people to participate and be a part of it is a big challenge, right? I think the opportunities are greater.Once you kind of get people rallying behind your side, a big part of this role is the change management piece of it. But I think we've demonstrated that we can do more with less by being smarter and more organized about how we approach the problem. I think this will lead to the next logical steps that might be improving documentation for billing inpatient documentation.Documentation on other services besides medical services. I think standardizing the data capture will be helpful for, you know, look at trauma as an example. Trauma's got a very specific expectation from the American College of Surgeons of what they want on their documentation standards. I think this is their right for the next opportunities.To then help with the data capture local facilities, there's a lot of trauma centers in our health system and they spend a lot of resources on that data capture for the ACS. So I think this will help them be more effective in their role and more accurate in their role.

Jordan Cooper   15:08
As we approach the end of this podcast episode, Josh, I'd like to ask you to reflect on the last few years of implementing these two centralized processes. And for the benefit of our listeners, do you have any advice that you would have given to your that you could give to yourself if you had the opportunity to speak to yourself before you went through these processes?

Joshua Moskovitz   15:26
Yeah, you know, I think patience is really important for this whole process and I think any expectations you have how quick something can go, you really need to slow that down even further. You may think it takes you six weeks, you make a bunch of 12 weeks, but you really want to give yourself 24 weeks. It's really with that very tongue and cheekly, but you know, the reality is you have to be slow, purposeful and.

Jordan Cooper   15:41
Uh.

Joshua Moskovitz   15:46
Make sure that you're sensitive to everyone's needs to really make this kind of level of change that we're working on here.

Jordan Cooper stopped transcription