Universal Health Services, Inc. (UHS) Earnings Call Transcript & Summary

Okay. So welcome back to the virtual component of the Stephens 2021 NASH Annual Investment Conference. I'm Scott Fidel, I'm the healthcare services analyst at Stephens. We're really pleased to host the next fireside chat today with Universal Health Services. UHS is one of the leading hospital management companies in the U.S., operating a highly scaled behavioral health business and an acute care hospital network, primarily concentrated in markets with attractive long-term patient demographics. Today, we're joined by Steve Filton. Steve's the Chief Financial Officer. Just want to note that for any investors that have questions that you'd like me to ask Steve, please email those to [email protected]. So Steve, thanks for participating again this year at the Stephens conference. Really appreciate you joining us once again.

Maybe just to start the discussion. Obviously, a lot of focus here on COVID and with the new variant having just hit the market. But I think before we sort of touch on that, just wanted to sort of pass the mic over to you to really sort of highlight just more broadly some of the key areas of strategic and operational focus for UHS right now as we're starting to put an end to 2021 and looking out to 2022.

Scott. So our last broad kind of public appearance was in our third quarter conference call at the end of October. At that time, we sort of reported acute care results, which I think reflected generally an outperformance in the third quarter and behavioral results, which reflected to some degree an underperformance in the quarter, both of which I think were tied to COVID volumes, which were quite high in the third quarter. What we also talked about at that time was that we had begun to see at the very end of the third quarter, kind of mid-September, a decline in COVID volumes that has continued, I think, at least until the last couple of weeks where we've seen COVID volumes in many of our markets, at least level off, in some cases, start to creep back up a little bit, but still at much lower rates than what we were generally experiencing in Q3. I think the most significant dynamic, at least challenging dynamic, in the third quarter was that increased COVID volumes created labor shortages and wage pressures in both business segments. I think on the acute side, those pressures tended to be outweighed by the higher revenue, the higher acuity created by the COVID patients themselves as well as largely a return to pre-pandemic levels of non-COVID activity in the emergency room traffic and in scheduled elective procedures. On the behavioral side, really not any sort of comparable offsetting of those costs and particularly those shortages. So we continue to experience muted volumes, maybe 5% off of prepandemic levels. What we expected at that time, 6 weeks ago or I guess at this point, almost 8 weeks ago, was that as COVID volumes eased, we'd see the labor pressures in both business segments ease in a commensurate fashion, although relying on our experience earlier in the year back in the January time frame, we expected that the sort of full impact and recovery from those labor pressures would take a good 3 or 4 months as it did back in the spring. And I think it's been playing out that way. So I think what we -- the way that we envision the year playing out back in the end of October has largely been our experience over the last 6 or 8 weeks. Now obviously, in the last week or 2, there's been much written and speculated about the Omicron variant and how that will impact things. And it's probably too early to tell. But I think absent that development, things have largely proceeded in Q4 so far the way that we expected.

Okay. Great. And I guess sort of picking it up on the Omicron here and certainly recognizing that this is very much a new emerging variant, and still a lot to be learned about whether this is going to have the same impact as some of the prior variants like Delta. I still think, Steve, we'd all be interested in your initial thoughts on that. And how, I guess, from a sort of initial operational planning perspective, UHS is starting to factor in the Omicron into your thinking?

Yes. Look, I think it is, again, too early to know whether -- number one, whether the Omicron variant will really create a surge in COVID patients, and if that surge will be materially different and have a different sort of characteristic feel to it than the Delta surge that we experienced in Q3. The good news about the Delta surge in Q3 was, particularly on the acute side was, we didn't see or experience, and I think this is true of our other acute care peers, of these public peers, that same sort of crowding out dynamic that we had seen in previous COVID surges. Our ER visits and our elective and scheduled procedures, as I said, were generally operating at pre-pandemic levels and have continued to do so again through the early parts of the fourth quarter. So that was the good news, and I think contributed to the acute care outperformance. On the behavioral side, I think like most other subacute providers what we saw during the COVID surge, was a migration of our labor force, particularly nurses, out of the subacute setting, and again, in our case, behavioral, but I've seen the same thing happening in nursing home businesses and home health businesses and other subacute businesses, where nurses in particular are leaving to work in acute care setting during these COVID surges. Whether that would sort of reoccur in Omicron surge and how significant that would be, I think, again, too early to tell. I don't think we're seeing that yet, and hopefully, we won't, but that would be the concern.

Understood. And then even before the Omicron, we have started to -- all of us are tracking. What's been a renewed uptick again just in I guess, more Delta-related COVID cases. I guess, November would probably be where there's been a bit more of that feature in the landscape again. Just interested, from your observations at this point, is this generally playing out the same way that the prior sort of initial phases of, let's call it, Delta we're looking for -- looking at that point? Or are there any notable differences as you think about impacts either on the acute or the behavioral side of the businesses?

Yes. Look, I think that, again, our expectation, and I'll be very candid in saying that as we've tried to predict the ebbs and flows of COVID volumes, I'm not sure that we've been any -- particularly any better than anybody else at this. I'm certainly not an infectious disease expert. I think some people have speculated that as the colder weather became more predominant and people began traveling for holidays, we'd see that same sort of surge that we saw post Thanksgiving, December and January of last year. On the other hand, obviously, a much more significant portion of the population was vaccinated this year than they were last year. So others have speculated it wouldn't be as grave. So far, again, I don't know that we've seen a really significant surge in COVID cases, we definitely have seen a leveling off, so they're not continuing to decline in number as they had for the last 6 or 8 weeks. But I think that's sort of where we're waiting to see what happens next. That we see COVID volumes come back up and create new labor pressures or have we just paused for a moment, and we'll continue to see a decline. And again, I just think it's too early to tell. And again, the Omicron variant sort of creates a whole new rub in all of this.

Got it. And then, you had mentioned the vaccines. Just wanted to get an update from you on that perspective just as we sort of move more into these federal mandates. And obviously, the acute care hospitals and even the behavioral hospitals, I think, had started to ramp up vaccinations of staff faster than a lot of other areas of healthcare providers, given a lot of the mandates even at the state level. But would just be interested in how the vaccination efforts on the staff side have continued to sort of play out for UHS and where things sit right now in terms of vaccination rates when looking at the acute care and the behavioral side of the business?

Yes. I mean, so I think that the vaccination rates of our employees have generally tended to reflect our discrete geographies. So if we're in a geography where vaccination rates are high, then it just seems like our employee vaccination rate was also high. I think generally, our employee vaccination rate has been a little bit higher than the overall population vaccination rate in whatever geography we're talking about. But -- so I think in a state like California where we've got a significant presence, I think we've got pretty high vaccination rates. In states like Arkansas, Mississippi, where we have behavioral facilities and the general population that's pretty low vaccination rates, I think our employees have higher than the general population, but still on sort of the lower side. Now we'll see what happens with the federal vaccination mandates. What our experience has been, we're in a number of geographies that have had their own vaccination mandates, State of California and New Jersey, Oregon, City of Philadelphia. What we've seen in those already -- what we've experienced in those geographies is, number one, as we get closer to the mandate date, vaccination rates do sort of pop up some. And then secondly, we just haven't had a significant loss of employees in any of those places. Small numbers of employees and ultimately, I'm not sure, honestly, what they decided to do in terms of other employment options, but they did leave the hospital. But they were relatively small numbers and not, I would say, or I wouldn't describe the impact as measurable on our overall operations. Whether that's going to be true with the broader federal mandate, I think it's still a little too early to tell, but at least that was our experience in the geographies that have already been through this.

Okay. Got it. So if we think going back to the 3Q call and during that, you had provided us with some preliminary thoughts around 2022, and how to compare that against 2019 sort of baselines. And just interested where we sit here today, basically at the end of November, early December. I mean obviously, Omicron could become a meaningful swing factor or maybe not. But just as you continue to see how the fourth quarter has continued to complete out, are any additional commentary that you want to make around that sort of initial 2022 framing that you provided in terms of headwinds or tailwinds or just sort of, I guess, how we're trending relative to those initial commentaries you gave us on the 3Q call?

So Scott, I think our thought process in giving that what we consider to be sort of very broad parameters of 2022 guidance was the notion that we would have some measurable amount of COVID cases in 2022. They would be at least as measured in our acute hospitals and -- as a percentage of our admissions kind of in the mid-single digits, as opposed to where they were in Q1 or Q3, where they were more in the low double digits, 12%, 13%, 14%. So that our results would resemble something a lot closer to Q2, than they would Q1 or Q3. That's still our thought. Again, harkening back to what we've already talked about. Feeling a little less confident that, that may be the case today than we did 6 weeks ago or 8 weeks ago, but I think that still has to be determined. But yes, I mean, it was sort of broadly based on the idea that if we're at more manageable level of COVID cases, not only in our geographies, but around the country, that would contribute to a significant easing of the labor pressures, both labor shortages and wage pressures, wage inflation. And our results, again, would look a lot more like growth over 2019 than it would over some of the ebbs and flows that we experienced in 2020 and '21.

Okay. And one thing, Steve, I was just thinking about the last day or 2 and just with Omicron and thinking about some of the impacts here. And obviously, there's a lot still that we have to learn around the severity and how much disease it causes and certainly transmissible, but how the vaccines play against it. But I think one thing it probably just reinforces is that the pandemic dynamics are going to just continue to sustain themselves likely well into 2022. And so one thing I was thinking in particular is when thinking about some of the policy backstops that have been put into place around the pandemic, and thinking about the public health emergency, for example. And just even seeing some of the initial government responses in the last few days, my suspicion here would be that the PHE, which I think, in general, the expectations have been that it would probably run through the middle of 2022, and then the question is whether it would sort of go longer than that. It feels to me like it's probably going to go longer than that, right, if I had to put a bet on this right now. But just interested in your thoughts around that front. I mean would you think it's reasonable that just the latest variant probably should serve as a reminder to all of us that, this framework that we're in right now with the pandemic, it's probably not going to be just a light switch changing in a quarter or two, right? It's going to take some time, probably for ultimately, all of this to manifest itself out.

Look, Scott, to be, again, perfectly fair, I don't know that we have any particular insight into what the government is likely to do. But I would generally concur with everything that you said seems to be reasonable and logical. And I think especially because what it points up, and I think you've alluded to this, is the difficulty in sort of predicting this. So for the government to do away with the Public Health Emergency designation after a couple of months of sort of lower volumes, sort of belies the fact that who knows what's going to happen with the variants, et cetera. So I think they're going to be more cautious about doing that than they might have otherwise been. So yes, I would guess, as you articulated that the Public Health Emergency and the additional reimbursement, sort of subsidies that have come along with that, are likely to be extended for some time into 2022.

Do you think that this -- obviously, this is total speculation, but do you think this gives a little bit more, I guess, sort of lobbying sort of point for the industry just around the Medicare sequestration side and sort of arguing that, this was clearly one of those key support mechanisms on the Medicare side of the reimbursement front that sort of shifting back to sequestration, effective Jan 1, 2022 could be disruptive, right, as we have another variant on the way. I mean obviously, it's speculation here, but any sense on just whether that type of dialogue, the industry may continue to try to sort of frame, whether it's the right timing here to really be sort of putting the sequestration cuts back into the Medicare reimbursement?

Yes. No, look, I think that same logic applies in the sense that while the Medicare sequestration waiver was never tied directly to the Public Health Emergency designation. It certainly was a response to the pressures and the financial pressures in particular, the COVID surges were putting on the hospital industry. And to your point, the industry is certainly, I think, making the argument to the government that those pressures are certainly not clearly been lifted, so that it would be prudent to keep the sequestration in place. I'm not political expert to really put a probability percentage on that. But I would say, it certainly seems to me those arguments are a lot more potent at the end of November than they were just even 2 months ago.

Got it. So as a sort of theme here, there's a couple of end markets that we're really trying to focus on around the Stephens Conference this week and really the 2 are the behavioral health care market and then the post-acute care market. Clearly, UHS plays very heavily into that first bucket. So I wanted to just spend a little bit of time with you here drilling into the behavioral side of the business, both sort of short term and long term. So I guess, first from the near-term perspective, you've been very clear and transparent around -- on the labor and staffing side for the behavioral business. And essentially, that's been creating the challenges in converting on very strong underlying patient demand into the volumes. So Steve, maybe we could just start there on the demand side of things. And if you could just talk to what you're seeing on the underlying demand side of behavioral services, particularly as we think about the reopening economy and some -- as we moved essentially to the next phase of the pandemic? I'm just interested in what your data has shown on whether you've seen any changes around the demand side first on the behavioral from, let's call it, sort of Phase 1 of the pandemic as we move into phase 2.

Yes. Scott, I think as you know, I mean, our commentary on behavioral demand has been pretty consistent throughout the pandemic. And it's really based on 2 things. I mean one is our own internal data and we track what we describe as sort of our inbound activity, meaning or what that encompasses are inbound inquiries, telephone calls to our 800 numbers, inbound, Internet inquiries, et cetera. And the volume of those inquiries has been doing nothing but rising during the pandemic. And I think, frankly, at certain points over the last 20 months, have risen to the level where those inquiries are 15% to 20% higher than they were pre-pandemic levels. None of that's changed. It continues to be high. I think it continues to be high, because I think behavioral demand had been on a sort of steadily increasing trajectory regardless or pre-pandemic to begin with. And then I think that the pandemic has created a level of sort of stress and anxiety, and complications to anyone, who is already having behavioral issues. How we've say to people, look at -- I think the pandemic has been stressful for the mentally healthiest amongst us, so you can only imagine if you were suffering from some sort of chronic behavioral illness, whether that's depression or addiction issues or whatever, that the incremental stress of the pandemic has only made that worse. And I think that's been our experience. We also look at macro data from any number of sources that again, suggest that really across the board, meaning all service lines, addiction illness, eating disorders, autism diagnoses, severe depression and suicidal ideation, all these things have generally been increasing, again, as measured by kind of broad macro metrics through the pandemic. So that, not necessarily from a social science perspective, but from the perspective of our business model, obviously, is all a good thing. The challenge that we've had, as you noted, and then we certainly have been probably the primary topic of conversation for the last 20 months, is our ability to meet that demand has been hampered by mostly our ability to fill all the vacancies that we had, whether those are nursing agencies or therapist vacancies or even what we call mental health technicians or nonprofessional people. What we have found during the pandemic is that our ability to fill those vacancies is really inversely related to the amount of COVID experience -- COVID we're expanding in our various markets. So during the COVID surges, that becomes more problematic. And I think it mainly becomes more problematic, and I'm guessing, Scott, if you're hearing this from other post-acute providers, because I certainly hear it from other post-acute providers, whether that's nursing homes or home health providers or whatever, that they're losing employees, particularly nurses, to the acute care setting. And there's always, as I said to be -- for as long as I've been doing this, which is several decades, there's always been a discount to nurses and other clinicians, who work in subacute settings. Again, home health, nursing home, behavioral, but that discount has been wildly exacerbated by the pandemic itself. So during the pandemic, you can go online and just if you Google nursing opportunities or whatever, you can see opportunities for nurses, who are working in a post acute or subacute setting to make 4 or 5x their salary, if they're willing to travel and work in a COVID environment that sort of thing. And as long as the nurse is willing to do that, there's really not much that we can do in terms of a countermeasure -- effective counter measure to offset that. So at one level, what I'd describe as sort of a temporary level, we just have to wait for those COVID volumes to ease and we're certainly doing that. But at another level where we think some of these changes may be more permanent, maybe more structural in nature, we are doing a lot of different things to change our patient care models, make them less reliant on our ends, more relying on other degree, lower degreed nurses that are easier to recruit and employ, or nonprofessionals like the technicians, et cetera, we're doing all those things. And back to a point that you made before in a different context, I think one of the things that will be helpful is that to the degree that the pandemic continues to kind of have a more omnipresent sort of effect on the regulatory environment, I think that regulators are going to be more amenable to and open to changing things like patient care models, because the alternative, and as we're having these conversations at state capitals and with state health authorities, et cetera is, if you don't change, if you don't give us some relief, if you will, on the RN staffing model, patients are going to go untreated. I mean that's really the equation here. So we're finding at least initially some receptivity on the part of regulators and policymakers to be open to these kinds of changes that maybe historically they haven't.

Got it. And yes, I said you clearly hit on a number of those -- the key dynamics that are playing out. And clearly, you have a lot of insight into this in terms of how operating both acute care hospital and behavioral businesses, how the economics of the pandemic essentially are just different for acute care hospitals in terms of the revenues that are generated by patients and how much as a result, acute care hospitals can pay the nurses relative to the impacts on the post acute and behavioral and other areas. And my follow-up question was going to be, and I think you sort of just answered it, but just as we try to look forward a little bit more, then I was going to ask you about what do you think ultimately will be the catalyst for creating some more equilibrium between supply and demand on the behavioral side? I mean, it sounds like you already addressed that, right, in some of your comments on the prior one, some stabilization in the COVID environment, some of these regulatory changes, perhaps. But I just wanted to give you the chance to further expand upon that in terms of your thoughts on sort of ultimately, when we -- whether it's from a timing or from a event perspective, we finally get to somewhat more of a balance here between supply and demand.

Yes. So look, I mean, people, investors all have been asking how much of these headwinds are temporary in nature and how much are more permanent, more structural in nature. And as with a lot of questions, I think the answer is somewhere in between. As I said, there's no question. I mean you can go back as again, as recently as the second quarter when COVID volumes were much, much lower than they are today, and in the third quarter. And our labor pressures were just not nearly as great and both of our business segments outperformed and we were well ahead of consensus estimates, et cetera. So you do have -- we're not just sort of doing wishful thinking here, you have a model to go back and look at, that's a relatively recent model, let's say, what does this look like in a kind of a lower COVID volume environment. On the other hand, people speculate that. Yes, I mean some of this, though, has changed the landscape permanently. Some nurses have retired early or just burned out and are not really wishing to work in a patient care environment or some nurses, younger nurses without families with more freedom are willing to work as traveling or temporary nurses for a more extended period of time to make kind of a premium wage rate. And I think all that's true. And again, one of the reasons why we're looking for permanent changes in our patient care models, because it sort of kind of assumes an underlying fact that RNs will be more difficult to employ and recruit than they have been historically. So I think we will get some relief from lower COVID numbers, but we're also actively trying to create this. And one of the points I made to somebody earlier today, and I think it's an important one is, not only are we changing patient care models to rely more on LPNs or certified nurse assistants or nonprofessional technicians, but as we hire these people, we're really trying to provide them a career ladder almost immediately, so we hire a tech and say, and by the way, tech, if you would like to become an LPN or a CNA or ultimately an RN, we're more than willing to invest in your education. And frankly, we'd encourage that, et cetera. So -- which -- it's a little bit different than maybe the opportunities that they may get elsewhere. And I think it's one of the things that a large organization like ours can afford to do that a smaller organization, a physician's office or something may not be able to do.

Understood. And then thinking about what how the payors sort of, I guess, play into this and how, I guess, this could influence contracting and just basically how even the structure of contracts, whether it's value-based or bundled or whatever. And you talked about some of the things on the regulatory side. Clearly, from the payor perspective, they've got an issue in terms of access, right, for their members. I mean I'd have to imagine a lot of payors have a lot of members right now that are struggling to get the access they need to appropriate behavioral care given the current backdrop. At the same time, having covered payors for as long as I have, we know that they don't want to raise the rate necessarily to have to -- to be able to improve the access dynamic. At the same time, it's felt like the payors generally probably relaxed some of the utilization controls, right, during the pandemic relative to prepandemic and there was at least some short-term assist from that. Not sure if that's continuing as much at this point or not. So sort of throw a bunch of things out there around the payors, but just interested in how they play into this ultimately. Because it doesn't serve them to have a system where a large percentage of their members cannot access the behavioral health that they need.

Yes. So look, I think it's worth noting, and I know people know this, but just sort of reinforcing the idea that 50% of our patients come to us roughly with some sort of government insurance, either Medicare or Medicaid. Neither of those programs are necessarily structured so as to sort of provide immediate relief to providers in an environment of high elevated wage inflation. And then in the commercial payors, as you point out, I mean, I think some investors have this view that we just automatically go to the payors and say, we are experiencing significant wage inflation, you're going to need to increase our rates and assume that the payors are willing and able to do that. Like you, that has not necessarily been my experience over the years. On the other hand, what I do see happening internally. And I think is that our approach and our sort of behavior in our managed care negotiation process is informed by the significant wage pressures that we're under. And even in the last few months, I feel like we're being more aggressive. Again, on very much a case-by-case basis, but as an example, if we're talking to a managed Medicaid payor, who says to us, we're not willing to give you an increase next year or in 2 years or whatever. And we're saying, well, we're experiencing 4% or 5% wage inflation as opposed to 2% or 3% prepandemic, without contractual price increase, we're willing to give you a termination notice or move out of network or whatever it may be. I won't say we're doing that across the board, but it certainly seems to me that we're more willing to do that today in a high wage inflation environment or higher wage inflation environment than we were 2 years ago in kind of a more modest environment. So it plays out, I think, in that way, Scott. And I don't know that there's an exact way of sort of saying, well, this is what it's going to ultimately mean to our contractual rates, and this is what it's going to mean to how many contracts were in network and not in network. But our ultimate leverage -- look, I always make the point that the ultimate leverage in any managed care negotiation is for the payor to say, we're not going to sign a contract and you're going to be out of network or for the provider to say, we're not going to sign a contract and we're going to be out of network. And depending on who has the greater leverage in the market and the greater market share and the availability of other opportunities for the payors, that I think affects how these things ultimately get resolved.

And as we think about perhaps converting that into some numbers and thinking about sort of financial trends or at least even sort of the composition of revenue growth behavioral business. I mean it certainly feels like more in the near term as a lot of this continues to take some time to settle out. The composition of revenue growth probably is -- remains more weighted towards price, right, than the pre-pandemic model, a little bit certainly less weighted towards volume. And then the question ultimately is sort of how that nets out to sort of sustain margin? And perhaps over time, whether it's through systemic change and structural changes, we can sort of regain that balance around the volume growth side of things, and start to bring down that pricing and so you get margin expansion going back on. But -- do you think that's sort of a reasonable way to think about it as we think about sort of the fact that none of what's happening right now is going to change overnight on the behavioral side?

No, no, no. So I think you've described the situation pretty accurately, Scott. Again, if people reflect back on how we used to talk about our behavioral model pre-pandemic, I think we used to think about a reasonable growth rate as being sort of mid-single-digit revenue growth same-store, but it was composed of price increases of 1% to 2%, and volume increases of 3% to 4%. During the pandemic, our price increases have been, in many cases, sometimes in that 6% to 7% range. I think what we've suggested to investors and just generally is that, look, we think some of that moderates as we emerge from the pandemic, but we have been getting contractual increases and they are sustainable. So instead of returning maybe to 1% to 2% pricing, maybe we land somewhere in the middle and sort of 3% to 5%. The good news in that regard is now we only need 1% to 2% volume growth to get to that sort of mid-single-digit bogey that we think is so critical to returning to EBITDA growth and margin expansion, et cetera. We're still running, as I was saying before, 4% to 5% below pre-pandemic volume level. So we're still ways away from even that 1% to 2% growth, but it's a lot easier to get there than against sort of 3% to 4% volume growth. So yes, I think for the foreseeable future, there's been kind of a remix, if you will, of our revenue components and skewed much more towards revenue and price growth than to volume.

And then the one other swing factor would just be around length of stay. And that's where we had been for multiple years prior to pandemic, we have been seeing consistent downward pressure on length of stay. I think Medicaid MCOs in particular, right, were probably a key driver of that trend in the system, particularly given how much behavioral services is funded through Medicaid. And then as I mentioned, sort of post pandemic there was a lot of relaxation, at least temporarily on the utilization, management controls from the payors. And maybe give us a sort of a real-time update on where things sit right now. I mean clearly, the managed care companies are concerned about MLRs and they're absorbing a lot of costs right now. The Medicaid MCO deal with some of the same types of dynamics in terms of their rates can often be behind cost trends, too. So just interested in sort of real time, how the utilization management dynamics, the interactions with the payors are on that and how you're going to -- how you're thinking about length of stay as a result of that over the near term?

Yes. So we saw, as you described, Scott, a pretty significant compression and pressure on our length of stay in 2017 and 2018. And one of the things that we said as we got into 2019 was, look, there's sort of a natural floor to this, I never really know who described it as the floor or ceiling, but sort of a natural limit to this, that what was really driving that compression length of stay was the shift in the Medicaid population from traditional Medicaid -- state-run Medicaid programs to managed Medicaid programs. And as more and more of our Medicaid population shifted to the managed programs, we said that those -- that length of stay compression and reduction would naturally sort of anniversary itself, kind of level off, et cetera. And I think we began to see that in earnest towards the end of 2019 and the beginning of 2020. And again, honestly, length of stay has remained very stable, during the pandemic has actually risen some, sort of gotten lost in all the other conversations, but I think we were getting there anyway. So I think as we emerge from the pandemic, I think we're generally -- our general expectation is that length of stay will likely remain relatively stable, that we saw the worst of that compression back in '17 and '18 as that population shifted. But now the vast majority of our Medicaid population is already in managed programs. So there's not that continued shift to be experienced.

Got it. I probably have time for 1 more quick one here. And just -- again, just as we think about some of the structural changes that can occur and what's sort of in the industry or in UHS' control. Maybe let's sort of leave it on that. One thing I want to ask about is just consolidation on the behavioral side is, is that something that could be helpful towards driving that equilibrium? Or do you think it's really -- not really a key relevant sort of driver of that? And then just anything else that you just want to sort of comment around this topic.

Yes. No, I think consolidation is an opportunity and a significant opportunity. The behavioral business remains a pretty fragmented business. We've talked a lot about the fact that more than half of the behavioral beds in the U.S. are run by or owned and run by acute care hospitals, who generally we think are not all that interested in that business, et cetera. So a lot of our growth we think, over the next several years, and a lot of the growth in the freestanding industry in general, will be through these joint ventures and partnerships with acute care hospitals for the beds -- that the behavioral beds that they control. And so I think that's a significant opportunity as the industry consolidates to a much greater degree. I think the other issue that is really unique to UHS is that, because we have a presence in both acute and behavioral, I think we have this unique opportunity to go to payors with our experience in both of the business segments and talk about and demonstrate, forget just talk about, but demonstrate the way that we can bring behavioral care to bear on acute care medical utilization experience. That's not something that, quite frankly, we've really done for most of our history. But I think there's a real opportunity, and I think payors will really value that opportunity. We had just started to go down that road when the pandemic hit. And so I think we really haven't -- had much of an opportunity to really sort of execute on that. But I think as the pandemic pressures ease, we'll really ramp up those activities. And like I said, I think they will resonate with the payors at the government level, at the private level because I do think we're in a unique position to really manage or help manage medical utilization for the payors/government/employer team.

All right. Great. Well, I'd like to chat about this for another hour, but I think we're out of time now. So Steve, again, I really appreciate you joining us once again at the Stephens Conference, not only for the fireside chat, but also for the investor meetings today. And again, just good to see you, and thanks for the time today.

Thank you, Scott. Good talking to you.