Encompass Health Corporation (EHC) Earnings Call Transcript & Summary

Good afternoon. I'm Crissy Carlisle, Chief Investor Relations Officer for Encompass Health. And it's my pleasure to welcome you to our first webcast of 2021, discussing how we use predictive analytics to reduce acute hospital readmissions and improve patient outcomes in our inpatient rehabilitation hospitals. I'm pleased to have Barb Jacobsmeyer, Executive Vice President and President of our Inpatient Rehabilitation Hospitals with me today. After a short video, we will transition to a Q&A session. For optimal viewing, viewers can maximize the video player to full screen. To exit, press the Escape button on your keyboard. [Operator Instructions] With that, let's get started. [Presentation]

It's now time to transition to the Q&A portion of our event. [Operator Instructions]

I'll kick things off with one, Barb. Tell us more about how you educate referral sources and payers on your use of these models and use of predictive analytics in order to drive admissions to our hospitals.

Well, Crissy, we share similar information to what we actually had in this presentation today in a slide format. But in addition, we actually bring Medicare claims data as it relates to that referral source or that payer's patient so that we can give them specific information regarding their patients, comparing their readmission data based on diagnosis, discharge setting. So for example, for our referral sources, we can actually use Medicare claims data to compare, let's say, their stroke patients and what their stroke patient's readmission rates are from our hospital versus maybe either other IRFs in the market or even the SNFs in the market.

What about our competitors? Do they have anything similar?

Not what we've seen. And it actually would be very difficult for them to have models like this because most of our competitors do not have one single electronic medical record, and it's been the valuable data from our electronic medical record that has helped us create these types of predictive analytics.

Here's one that I'm sure everyone's interested in: How does all of this translate to the bottom line?

Well, a couple of things. One, our reimbursement at our hospitals are based on a patient's CMG, or case mix group, that's comparable to an acute care hospital's DRG. We receive the full CMG payment when a patient discharges back to the community or back home. That drops to a short-stay or a per diem rate if the patient transfers acute or goes to a skilled facility. So for example, if the expected length of stay for a certain diagnosis is 10 days and the patient returns to the acute hospital on day 4, we would receive approximately 40% of that expected payment.

A.J. Rice with Crédit Suisse is asking, "What is the timeframe for rolling these predictive models out to your entire portfolio?"

Well, for the ReACT, ReACT was rolled out a couple of years ago. For this new readmission prevention data analytics, we actually rolled that out to all of our hospitals by the end of 2020. It does take a while to pilot these types of things because if you're not sensitive enough in the preventative data analytics, then you end up with too many patients hitting the variables, and then it's hard to have actions related to that. So it does take, like we mentioned on the video, it took 2 years to pilot the readmission prevention program so that we could get it sensitive enough where it's valuable enough to use in our markets. But once we have those, it usually takes about 6 to 8 months to get it rolled out to all of our hospitals.

A.J. also asked, "Are the results different at JV hospitals versus your wholly owned hospitals?

No, not necessarily. In the JV hospitals, we still get actually a lot of admissions from outside our actual JV partners. So the types of patients we get are comparable, whether it's a JV or a non-JV hospital. And so the outcomes are usually pretty consistent.

Barb, tell us a little bit more about how a patient transfer negatively impacts our quality and reimbursement.

So for quality, we're benchmarked actually with other inpatient rehab hospitals, and that's usually benchmarked on discharge status outcomes. So returning as many patients back to their home and/or community is the best outcome. If our acute care transfers are high, that impacts, obviously, the percentage of the patients going home successfully. And so it's important for us to show well as we're compared to other inpatient rehabs in our markets. And then as I mentioned, our reimbursement is based on the patient's CMG. And so we kind of touched on our reimbursement, but it also does impact the financial performance of an acute care hospital. Some diagnosis impact the acute care hospitals readmission penalty. All returns to acute impact to hospitals Medicare spend per beneficiary. And many hospitals are participating in either accountable care organization or bundled payment initiatives. And in those situations, it actually impacts their target spend. As I mentioned on the video, an average acute care transfer costs $14,000. That means each time a patient goes back to an acute, an average of $14,000 is spent towards that target spend that could potentially be avoidable.

[Operator Instructions] Barb, Kevin Fishbeck of Bank of America is asking, "Medicare Advantage has been growing rapidly for you. Do these capabilities help you get contracts or better rates?"

In many of our contracts, it does help us with our rates. Some of our annual increases are dependent upon our quality outcome, and so improving on these quality outcomes does help us as we negotiate each year on those increased rates with the Medicare Advantage plans.

And Barb, Matt Larew with William Blair asked a similar question: "Have you had success using the improved outcomes from these models to drive market share gains, preferred partner status or better commercial pricing?"

I would say yes to all of those. I mean each one of these, though, is an individual meeting with a referral source, whether it be an acute care hospital, a physician or a payer. And it's bringing this type of information so that they know -- even when we share our outcomes today, we feel like it's important for them to know what things we continue to focus on so that those outcomes continue to improve in the future.

Barb, here's another one from A.J. Rice: "When it comes to interacting with the home health provider, is there a difference between your experience with Encompass Home Health and other home health providers?"

It's a great question. What I would say is our relationship and clinical collaboration with our home health has been critical in developing these programs. It's how we use the data to determine the predictive analytics. It's also how we work to develop the best practices. However, after we completed the pilot, and we made sure we did those in markets where we were overlapped so we could work with our home health, we did roll it out to all of our markets, including our non-overlap markets. And it's taught us how to work well with all the home health, whether it's our home health or another.

All right. [Operator Instructions] Barb, what types of acute care transfers are necessary and unavoidable?

Well, some examples would be, we have a patient admitted that had a stroke, and it's not uncommon for some stroke patients to actually have a second stroke. In that situation, that would be unavoidable, and that patient would go back to acute. Another example would be a patient that has some sort of respiratory failure. We do not have ventilators in our hospitals. So if a patient gets to a point in their respiratory status where they need a mechanical ventilator, that would require a transfer. And then certainly, a patient -- as we talked about the importance of limiting fall, there are times that a patient falls and that does result also in a need for an acute care transfer. So those are a few examples.

Barb, Larissa Heywood with ClearSky Health asks, "Can you expand some more on the methods used to prevent acute care transfers?"

Sure. So when you look at the data that we have, so for example, if a patient is trending from that low risk to a moderate or high risk, we can actually click on that particular patient's ReACT score to understand what changed in those clinical variables to have that patient move from a low risk to a high risk. An example could be we may see that a patient's blood pressure has gone up. And so if a nurse notices that they move in their risk for the ReACT or readmission and they can drill down and see that it has to do with that patient's blood pressure, they can quickly reach out to the physician and get an adjustment, perhaps in the patient's medication or in their diet or something that would help control that so that we could take care of it in our hospital and avoid an acute care transfer.

Barb, tell us some more about the clinical variables that are different between ReACT and the readmission prevention model.

So when you think of ReACT, that is while the patient is in our hospital. And so most of the time, the patients are getting lab work done daily; sometimes, a couple of times a week. And so we're able to see what's happening with their lab work. When a patient returns home, that's no longer going to be available. And so there are things that you have to use differently when a patient is going to be transferring home. So when we look at the readmission prevention model, things like the patient's medications are going to be important. For example, inside the hospital, if they are diabetic, we're monitoring their blood sugars on a regular basis. But when they go home, we're relying on the patient to do this. So it's going to be important that we know what type of medication they're on. The patient's weight is a great example. Inside the hospital, they -- if a patient has congestive heart failure, they're being weighed on a daily basis. That, we're going to rely on the patient to do when they go home. Social determinants are an important part of this. The patient's access to transportation, family support, financial resources for things like their medications and food are critical components of the readmission prevention model for when they are returning home.

Barb, Scott Fidel with Stephens asks, "How have your predictive analytics helped to drive financial performance in value-based care programs with managed care MA payers?"

I think the best example would be when you think of the programs, the initial thought for many of these programs on a value-based was the thought that the skilled nursing facilities would be the best setting really because of their lower cost. And what these programs have helped us show -- and again, we're not using all our data. We're also using Medicare claims data to show that when a patient comes to an inpatient rehab hospital, if we're able to get them home and prevent these readmissions, in the long run, the higher price tag of an IRF stay may outweigh what the cost -- or what the benefits would be had the patient gone to the skilled facility.

All right. We'll give you just another second to see if anyone else submits a question via the Ask a Question. Barb, what about this? Would you ever license ReACT to other IRFs?

I think the short answer would be no. And the main reason really is because it's not just the data analytics program that's critical, it's the electronic medical record running behind it that's so important. And so folks need not only the access to the tools, but access to the electronic medical record for support.

All right. We're seeing no other question's coming up in the chat box. On behalf of Encompass Health, I thank you for spending time with us today. This is the first webcast of this kind that we've done. And tomorrow, you will receive a short online survey regarding your participation in today's event. I hope you will take a few minutes to complete that survey and provide constructive feedback to us so that we can continuously improve our communication with you. Thanks again for joining us.

This does conclude today's Encompass Health's Predictive Analytics webcast. Please disconnect at this time.